JOURNAL

American Society of

Sugar Cane Technologists

Volume 9 Florida and Louisiana Divisions

December, 1989

ASSCT

OFFICERS AND COMMITTEES FOR 1988

General Officers and Committees

General Secretary-Treasurer Denver T. Loupe

Program Chairman Barry Glaz

Executive Committee J. W. Beardsley Harold Birkett Martin Cancienne Tirso Carreja Frank Coale Ron DeStefano Humberto Farinas Steve Guillot, Sr. Ben Legendre Denver T. Loupe Lowell L. McCormick Charles Savoie, Jr. Omelio Sosa, Jr. Dale Stacy Roland Talbot Jackie Theriot Luis Zarraluqui

Editors of Journal

Managing Editor

Lowell L. McCormick

Technical Editors

Agriculture

Fred A. Martin

Manufacturing

Stephen J. Clarke

Florida

Tirso Carreja Dale Stacy Ron DeStefano Omelio Sosa, Jr. Luis Zarraluqui Humberto Farinas J. Wayne Beardsley Frank Coale

Divisional Officers

Office

President 1st Vice President 2nd Vice President

Chairman, Agricultural Section Chairman, Manufacturing Section

Chairman-at-Large Immediate Past President

Secretary-Treasurer

Louisiana

Roland Talbot Charles Savoie, Jr. Martin Cancienne Steve Guillot, Sr. Jackie Theriot Ben Legendre Harold Birkett Lowell L. McCormick

TABLE OF CONTENTS

Page

1 President's Message - Florida Division Tirso M. Carreja

2 President's Message - Louisiana Division Roland Talbot

6 Evaluation of Traits Associated With Resistance to Sugarcane Smut Caused by Ustilago Scitaminea C. P. Chao, J. W. Hoy and F. A. Martin

17 The Production of Seedlings in the Louisiana, "L", Sugarcane Breeding Program Keith P. Bischoff, Joey P. Quebedeaux, and F. A. Martin

22 Economies of Size in the Louisiana Sugarcane Processing Industry Brian A. Chapman and Ralph D. Christy

30 A Method for Determining Spore Production of Sugarcane Rust, Puccinia melanocephala James M. Shine, Jr., Victor Chew and J. D. Miller

38 Crop-Herbicide Management Options for Johnsongrass Control in Fallowed Sugarcane Fields Edward P. Richard, Jr. and Howard P. Viator

44 Sugarcane Response to Mn Sources and S Application Grown on Two Florida Histosols David L. Anderson and M. F. Ulloa

52 Effect of White Grub Ligyrus subtropicus (Blatchley) Infestation on Sugarcane Root: Shoot Relationships

F. J. Coale and R. H. Cherry

56 The Response of Sugarcane Selections to Sugarcane Borer in the Greenhouse and Field W. H. White and J. W. Dunckelman

62 Family Performance at Early Stages of Selection and Frequency of Superior Clones From Crosses Among Canal Point Cultivars of Sugarcane

P. Y. P. Tai and J. D. Miller

71 Yield Effects of Sugarcane Smut Infection in Florida B. Glaz, M. F. Ulloa and R. Parrado

81 Nutritional Status Survey of Sugarcane in Texas J. R. Thomas and N. Rozeff

91 Growth Response of Six Sugarcane Cultivars to the Herbicides Asulam, Dalapon, and MSMA R. W. Millhollon and Hugh P. Fanguy

97 Efficiency of In Vitro Propagation of Sugarcane Plants by Direct Regeneration From Leaf Tissue and By Shoot-Tip Culture

Michael P. Grisham and Druis Bourg

i i

AGRICULTURAL ABSTRACTS

Page

103 Effects of By-Product Gypsum on Yield and Nutrient Content of Sugar Cane and Soil Properties J. A. Breithaupt, Allen Arceneaux, and Ray Ricaud

103 Correlation of Crop Age With Populations of Soil Insect Pest in Florida Sugar Cane Ron Cherry

104 Ratoon Stunting Disease Losses in Four Commercial Sugar Cane Clones in Florida J. L. Dean, M. J. Davis and N. A. Harrison

104 A Progress Report on Harvesting System Performance at Two Florida Sugar Mills B. R. Eiland

105 Performance of the New Sugar Cane Variety CP 79-318 in Infield Variety Tests Hugh P. Fanguy

105 The Role of Stalk Density, Pith and Tube in Sugar Cane Selection K. A. Gravois and F. A. Martin

106 Some Studies on Damage to Sugarcane by the Spider Mite Otigonychus Stickneyi (McGregor) David G. Hall

106 Role of Varieties, Weather Conditions and Management Decisions in Record Sugar Yields for Louisiana in 1987

Benjamin L. Legendre

107 Improving Sugar Cane Varieties - What Are Our Options? F. A. Martin, S. B. Milligan, K. A. Gravois, and K. P. Bischoff

107 Inbreeding in the Louisiana Sugar Cane Variety Development Program and the Utility of Inbreeding Coefficients and Pedigrees in the Variety Selection Process

Scott B. Milligan

108 Influence of Propiconazole on Emergence of CP 74-2005 Richard N. Raid

108 Selection for Sugar Cane Borer Resistance in Seedling Progenies W. H. White, J. W. Dunkleman, and B. L. Legendre

MANUFACTURING ABSTRACTS

109 Turbidimetric Evaluation of Novel Clarification Schemes and Evaporation G. A. Adongo and S. J. Clarke

109 Parameters for Vacuum Pan Automation G. L. Aleman

109 Short Term/Long Term Computer Application for the Sugar Industry E. Alfonso and R. Valdes

i i i

Page

110 Evaluations of the Performance of a Forced Feed Roller on the Seventh Mill at Atlantic Sugar Association

J. F. Alvarez, H. Cardentey, and A. Pacheco

110 Computer Model to Assess the Economic Value of a Sugar Cane Variety S. J. Clarke and S. B. Milligan

110 Crown Wheel Removal from Bagasse Roll G. Delaune and J. Theriot

111 Microprocessor Control Structures for Raw Sugar Factories W. Keenliside

111 Monocast Nylon Mill Bearing Liners K. McGrew

111 Improving Performance of Low-Grade Crystallizers Y. Oubrahim, M. Saska, and M. Garcia

111 Cane Knives Choke Protection A. L. Perera

112 Preparing Cane with an Electronic Governor L. R. Zarraluqui

113 Editorial Policy

115 Rules for Preparing Papers

117 Author Index

iv

PRESIDENTS MESSAGE - FLORIDA DIVISION

Tirso M. Carreja Sugarcane Growers Cooperative of Florida

Belle Glade, Florida

The Florida sugar industry has completed the best crop in its history. Helped by two consecutive years without freeze and in spite of a drought in the middle of the cane growing season, the industry imposed the following three new records for Florida: most sugarcane ground, most raw sugar produced and highest sugar yield from sugarcane, all for one season.

The crop season began October 20 and ended March 22 with a total of 155 crop days. In that period of time, we ground 13.74 million short tons of sugarcane, produced 1.51 million short tons of sugar raw value, with a yield of 10.84 percent. We also produced 90.14 million gallons of final molasses 79.5 Brix.

Florida has kept the pace as the number one sugar producer in the United States thanks to the dedication and hard work of individual people and organizations; from agricultural researchers, to the executives that make the final decisions. All of them working together have helped to achieve this goal.

To sell and ship the raw sugar and final molasses produced, the Florida sugar industry depends on two cooperative organizations: The Florida Sugar Marketing and Terminal Association, Inc. and the Florida Molasses Exchange, Inc.

The Florida Sugar Marketing and Terminal Association, Inc. was formed in 1978 by five raw sugar producers, which built a deep-water loading facility at the Port of Palm Beach, with an initial expectation of shipping 250,000 tons per year to markets that could not be economically reached by rail or truck. With the completion of this terminal in 1979, the Florida industry was able to sell to all U.S. refiners and trade houses. This, along with increased sugar storage capacity enabled the processors greater flexibility in achieving better sugar prices as they were no longer restricted to dealing with a limited number of buyers who insisted on discounts to the market.

In recent years, this facility has also enabled the Florida industry to resist some of the onerous quality terms imposed by some refiners, in the form of the revised dextran and color tests.

The success of the terminal is best measured by the volume actually shipped through it. Quantities have increased each year, and by the end of 1987 the Florida industry had shipped 5,486,571 tons in nine years of operation, culminating with just over one million tons loaded out in calendar year 1987 alone.

The Port of Palm Beach the Florida Molasses Exchange, Inc. also operates, on behalf of all the Florida sugar industry, a deep-water molasses export terminal that has grown from 6,500 tons holding capacity at construction in 1973 to 40,000 tons currently, with plans to expand to 60,000 tons during Autumn, 1988. Approximately half of Florida's 500,000 tons molasses production is moved by highway semi-trailer to the terminal for subsequent shipment to feed grade molasses buyers in Europe and Canada. In conjunction with distributor's highway trailers and a combined fleet of over 100 jumbo rail tank cars, the port terminal enables the Florida industry to maintain a balanced marketing program.

We are also members of the Florida Sugar Cane League, and rely upon it to assist the sugar industry in subjects related with environmental, legislative, public relations, agriculture research and others subjects common to the industry.

During the past year, the Florida sugar industry has addressed many concerns regarding Florida's environment. The industry, through the Florida Sugar Cane League, formed the Environmental Quality Committee in 1968 in order to address both air and water issues, and has had to continually call upon its member's expertise to protect agricultural interests.

Water quality and quantity issues have been in the forefront. With the passage of the Surface Waters Improvement Management Act (SWIM) in the 1987 state legislature, emphasis has been directed to Lake

1

Okeechobee and the downstream impacts of runoff on the water conservation areas and Everglades National Park. The industry has taken the lead in the development and support of research on air and water quality in the Everglades Agriculture Area, Lake Okeechobee and surrounding areas. We will continue to play an active role in finding solutions to balance the water needs of South Florida's fragile ecological system.

In an effort to raise funds needed to conduct further water quality research, growers in the Everglades Agricultural Area have volunteered to tax themselves by forming a special taxing district. Although we have encountered many bureaucratic roadblocks, we are confident that our proposal will become a reality.

In addition to water programs, the industry conducts one of the largest private air monitoring networks in the state. The industry is also experimenting with more advanced monitoring such as PM-10's and a dichotomous sampler that be used in the future.

Presently the sugar industry looks good. We are almost in the middle of a five year sugar legislative program that allows the sugar industry to survive, but at no cost to the government. We have to remember that. Now if we look back ten years and evaluate the industry and market trends, we must start worrying about the near future.

In 1978 the U.S. market size was 14.0 million tons of nutritive sweeteners. Sucrose's share was 10.2 million tons (refined basis), with 5.5 million produced from cane and beet and 4.7 million from offshore suppliers. The other 3.8 million went to corn sweeteners and various nutritive sweeteners.

In 1987 we find a very difficult situation. Total market size is 16.1 million tons of nutritive sweeteners. Sucrose share is only 7.6 million tons (refined basis); 6.6 million from cane and beet, and 1.0 million from offshore suppliers. The other 8.4 million went to corn sweeteners and various nutritive sweeteners.

As we can see, total nutritive sweeteners consumption in the U.S. has increased 2.2 million tons in ten years, while the sucrose consumption has decreased 2.6 million tons in the same period. This mans, that not only did we lose 2.6 million tons of the existing market, but also that we did not receive any benefit from the growth in the market.

In 1987, for the first time in the last ten years, the sucrose consumption increased compared with the year before, and it looks like liquid corn sweeteners have reached market maturity. But the corn sweetener producers are working very hard developing a crystalline high fructose to compete in price and quality with sucrose. That is a threat that we can not ignore.

Until now domestic production hasn't been affected. We have increased our sucrose production by 1.1 million tons in spite of 2.6 million tons of decrease in consumption. The 3.7 million tons net decline have been a loss to the offshore quota. But the offshore quota is almost drained out, and if the present trend is followed, by 1991 we may have no tool to effectively administer the existing sugar program as well as making it more difficult to achieve a new sugar program in the next Farm Bill.

We have to prevent this situation from becoming a reality. We have to work to increase the sucrose consumption and control the sugar marketing. I know that it is something very easy to say, but very difficult to carry out.

The best way to increase the sucrose consumption is by using advertising and promotion. We are living in a marketing society and the industry has to become more marketing oriented. We know all about that, because the sugar industry has been under attack for many years using the myth that sugar is dangerous to our health. This myth has gone so deep in the mind of the people, that many manufacturers advertise their products with slogan "sugar free". It is time to go back to the people and tell them the truth; that sucrose is pure, 100 percent natural and the "gold standard" in sweeteners. We are now in a competitive market where we have to advertise our product if we want it to survive. The advertising is expensive, but it will yield dividends.

Besides advertising, we have to control the domestic sweetener marketing in the country. We cannot continue the present trend of increased production in a limited market. We have to work hard together with other nutritive sweetener producers to come to an agreement, and work on a legislative program that will limit domestic sweetener marketing in the country. If we don't do this now, in the future we will produce more sugar

2

than is consumed in the country leading to a loss of the Sugar Program. And the sucrose "below production cost" dumped by foreign sugar producer countries, will come directly into the U.S. market and will deeply hurt the U. S. sweetener producers. I know that many of you do not like to hear about marketing quotas, but somebody, someday must take the first initiative.

Think about it, but please do not wait too long.

3

PRESIDENT'S MESSAGE - LOUISIANA DIVISION

Roland Talbot Ronald Talbot Farms Thibodaux, Louisiana

On behalf of the Louisiana division of the American Society of Sugar Cane Technologists, I want to thank the Florida division for hosting the Eighteenth Annual Joint Conference in Clearwater, Florida.

The 1987 Louisiana sugarcane crop began where the 1986 crop ended: that is, fields having deep ruts filled with water from the excessive rains that fell throughout the harvesting season of the 1986 crop. The first priority was to drain the fields as soon as possible. The cane belt experienced several days with either a light freeze or frost which resulted in the loss of some terminal buds. Rainfall delayed much of the field work up to mid-April when fields became dry enough to begin to cultivate, apply fertilizer and spray for weed and grass control.

The 1987 crop produced 6,675,000 short tons of sugarcane and was processed into 740,000 tons of raw value sugar. The 265,000 acres of cane were delivered to 21 mills before Christmas. The Louisiana mills grinding operations lasted an average of 64 days, compared to 70 days for 1986. There was 11 percent less cane produced and 11 percent more sugar processed in 1987 than 1986. When warehouse sugar inventories are finally delivered to refiners, this will be either the second or third largest sugar crop produced in Louisiana. Mill production average for raw value sugar per ton in 1987 was 222 pounds compared to 181 pounds for 1986, an increase of 23 percent.

Climatic conditions for the 1987 crop had a significant bearing in the record high yield. The Louisiana Gulf Coast was never threatened by hurricanes and the weather fronts that came through the cane belt were generally not violent. The results of these conditions were an erect sugarcane crop which made harvesting cleaner cane a pleasant experience.

Comparing records of climatic conditions for 1986 and 1987 from the USDA Experiment Station in Houma should help answer the question: Why a 23 percent increase in sugar per ton yield? Rainfall during the months of May and June of 1987 totaled 23.6 inches compared to 11.4 inches for May and June of 1986, or a 106 percent increase in rainfall for 1987. For the same period of 1987 there were 39 days with rainfall compared with 27 days in 1986, or a 44 percent increase in days of rainfall for 1987. These conditions prevented some growers from completing their fertilization operations and, in most cases, prevented the completion of cultivation and layby operations. These wet conditions and reduced sunlight resulted in a reduction in the number of tillers which were able to survive and turn into millable stalks.

The harvesting season also had climate conditions which contributed to the record yield. Rainfall recorded for October through December 1987 was 8.1 inches compared with 1986 which was 15.8 inches, or a decrease of 48 percent to 1987. Low humidity readings for the three harvesting months for 1987 averaged 42 percent compared to 76 percent for 1986.

The cooler and dryer weather conditions of September and October resulted in cane maturing sooner than normal. Growers treating cane with Polado also experienced increases in sugar per ton over untreated cane. The lower humidity resulted in a more complete burn of cane leaves and grasses. With drier fields, little or no mud was delivered to mills. Little or no dextran problems were encountered due to the cooler and drier than normal conditions. Each of these climatic conditions had its effect on the 1987 crop. The results was the phenomenal record of producing more pounds of sugar per ton than anytime in the 150 year history of the Louisiana sugarcane industry.

Through the efforts of the our sugarcane variety research program, a new cane variety, CP 79-318, was released to the producers in 1987. This variety was jointly developed and released by the cooperating agencies, the Agricultural Research Service of the U.S. Department of Agriculture, the Louisiana Agricultural Experiment Station of the Louisiana State University and The American Sugar Cane League. According to scientists from the three agencies CP 79-318 should yield as well as CP 70-321 and CP 65-357 in both tonnage and sugar per ton.

4

The increased use of the two-row harvesters and loaders with mechanical cane pilers has helped contribute to the outstanding sugar per ton yield. The development and use of the two-row loader will further enhance the quality of cane delivered to the mills in the future. Most of the cane planted in Louisiana is now planted with mechanical cane planters, even though efficiency has yet not been achieved. Cane planters basically have two problems areas that are in need of engineering research. More cane is planted than necessary and damage to seed cane needs to be reduced. Many mechanical planters are planting in excess of 56,000 eyes per acre with approximately four percent damaged by cutting and loading operations and approximately eight percent damaged by mechanical planters. Planting over 30,000 eyes per acre does not increase final stands or yield. If mechanical planters were more efficient, there could be a 40 percent decrease in cane used for seed.

Processors, cooperating with the Audubon Institute, will have to be more cost-efficient, reduce loss time by improving equipment performance and with computer technology, automate mill operations when practical to eliminate human error. Research is needed to develop a quick, reliable test for dextran in cane juice and methods for dextran control and elimination.

The Reagan Administration has not given up its desire to change the sugar program by lowering loan rates from 18 to 12 cents per pound. With import quotas reduced to 750,000 tons and blended sugar approaching 900,000 tons, entering the United States outside the import quotas system, our successful "no-cost" sugar program is threatened. Organizations represented by some of the largest sugar corporations in America are opposing the sugar program. Only the self-interest for profit is the justification for their opposition.

The challenge in the future will be to preserve a viable domestic sugar industry. The first challenge will be to prevail on the members of congress and assure the consumers that sugar can be purchased at reasonable, stable prices and changing the sugar program will not be in their best interest. Many consumers believe that if no sugar program existed they could purchase sugar at the so-called "world price" of eight cents a pound. That myth would soon turn to a deplorable reality when sugar prices would escalate without a domestic sugar industry.

We will have to be more aggressive in marketing our product. Competition in the sweetener industry is fierce and backed with advertisement programs costing millions of dollars. Numerous consumers also believe that sugar is harmful and should be not be consumed. Sugar is a safe product. It is the highest quality sweetener, pure and 100 percent natural. It is long past time for our industry to change its thinking. We must change from a passive defense to an active offense before its too late. Now is the time for the sugar industry to seize the opportunity to work collectively to market our product.

The people of the Louisiana sugarcane industry have survived because of the determined character of the people that make up this industry. Adversities are common, they range from the freeze that happens too late in the spring or too early in the fall. The rains that fall when a drought would be more beneficial, a drought when the crop is withering for need of rain, high winds and rain that flatten cane to the ground, etc. Nevertheless, after each year we bounce back with renewed enthusiasm ready to produce another cane crop.

The Louisiana sugarcane industry is grateful to those, past and present, who have helped and supported our industry by participating at joint meetings of the American Society Technologists. Without the support of our universities, dedicated personnel of the USDA, Louisiana State University, The Cooperative Extension Service and the American Sugar Cane League, the knowledge needed to survive in an ever-increasing competitive sweetener market, could be the difference between survival and extinction. This industry would certainly not survive if it were not for the unanimous support of our congressional delegation and our alert lobbyists. Through the effort of the American Sugar Cane League, congressmen from other states have visited our industry and left with a better understanding of the need of having a viable domestic sugar industry for the good of the taxpayer, the consumer and the industry.

5

EVALUATION OF TRAITS ASSOCIATED WITH RESISTANCE TO SUGARCANE SMUT CAUSED BY USTILAGO SCITAMINEA'

C.P. Chao and J.W. Hoy Plant Pathology and Crop Physiology Department

and FA. Martin

Agronomy Department LSU Agricultural Center

ABSTRACT

Resistance to smut, caused by Ustilago scitaminea,was evaluated in 15 hybrid clones of Saccharum with smut reactions ranging from resistant to highly susceptible. No consistent association was found between smut infection level and bud traits including length, width, length x width, shape, flange, groove, germination type, germination times, and shoot growth rates. In two clones, CP 65-357 and CP 74-383, inoculation of plants with initial shoot lengths up to 6 cm resulted in high smut infection levels, whereas inoculation of ungerminated buds or plants with longer shoots resulted in low infection levels. Resistance to systemic infection was detected in plants of resistant and susceptible clones following wound inoculation.

INTRODUCTION

Sugarcane smut, caused by Ustilago scitaminea Syd., is an important disease of interspecific Saccharum hybrids worldwide (1). Infected sugarcane stools contain grass-like, unmillable shoots which produce a long, unbranched, terminal "whip" at the apex. Billions of fungal spores are produced and released from each whip (16). Yield losses usually increase with successive ratoons and can be very severe in susceptible cultivars.

Resistant cultivars have proven to be the most effective measure to reduce the incidence and severity of smut. Hence, breeding and selection for smut resistant clones receives major emphasis in sugarcane breeding programs wherever the disease occurs (4,5,8,9,18). Smut resistance levels are typically evaluated by dip-inoculating stalks of clones in smut spore suspensions, planting the stalks, and comparing the levels of infection that occur in different clones during the growing season. Repeated trials are necessary to reliably determine and evaluate clone smut reactions. This is a time consuming and expensive process requiring large amounts of space. Therefore, research has been conducted to determine if any plant characteristics are consistently associated with resistance to smut (2,11,12,17).

Several bud morphological traits and growth characteristics have been reported to be associated with susceptibility to smut (11,17). These traits include large bud size, triangular bud shape, absence of a bud flange, presence and depth of a bud groove, apical type of bud germination, rapid bud germination rate, and rapid initial shoot growth rate. In addition, research has attempted to determine shoot lengths at which germinated buds become resistant to infection (3). It has been suggested that bud traits are variable under different environmental conditions (8,11,17); therefore, confirmation of these associations with more cultivars in different environments is needed.

Most of the research to determine the nature and mechanisms of expression of smut resistance has focused on bud characteristics or resistance to infection. However, consistent differences in the expression of disease between cultivars have been observed, and different types of resistance to sugarcane smut have been postulated (6,7,10). A pre-infectional type of resistance revolves around a barrier to initial infection represented by the bud scales (2,12). Post-infectional types of resistance may prevent the establishment of infection and affect the extent of fungal colonization and disease expression (6,10).

'Approved for publication by the Director of the Louisiana Agricultural Experiment Station as manuscript number 88-38-2404.

6

The first objective of this study was to evaluate sugarcane cultivars in the Louisiana breeding population for the association of smut resistance with selected morphological traits and growth characteristics. The second objective was to evaluate cultivars for the expression of what has been termed post-infectional resistance to smut (7,10).

MATERIALS AND METHODS

Fifteen sugarcane clones selected from the Louisiana sugarcane breeding population, L 65-69, CP 65-357, CP 66-346, CP 67-412, CP 70-330, CP 72-355, CP 72-356, CP 72-370, CP 73-308, CP 73-351, CP 74-383, CP 76-340, CP 77-310, CP 77-407, and CP 77-413, were chosen to study the association between bud morphological traits, germination, and growth parameters and resistance to smut.

Smut ratings of tested clones

During September 1985, three 6-stalk replicates of each clone were dipped for 10 minutes in a freshly prepared smut spore suspension containing 5x10 spores/ml. The three replicates of each clone were planted in a randomized block design in single row plots 2.7 m in length with 0.9 m alleys between plots.

The number of smut-infected stalks and the total number of stalks in each plot were recorded nine months after planting. The overall smut infection percentage of each clone was calculated by averaging the smut infection percentage over three replicates. The level of smut resistance in each clone was rated on the basis of comparisons of overall smut infection levels with infection levels in clones with known smut reactions (CP 72-356, resistant; CP 65-357, moderately susceptible; and CP 73-351, highly susceptible). Based on these comparisons, clones with an overall smut infection percentage less than 11% were rated as resistant (R). Clones with smut infection percentages ranging from 11-22% and greater than 22% were rated as moderately susceptible (MS) and highly susceptible (HS), respectively.

Bud morphological characters

During October 1986, 40 nodes in the middle portions of four stalks of each of 15 clones were chosen for measurements of bud length, width, and shape and observation for bud germination type and the presence of a bud flange and groove. Bud length x width was calculated and used as an approximation of bud size. The degree of correlation between bud length, width, length x width and the inoculation test smut infection percentage was determined by regression analysis. Since bud shape, groove, flange, and type of bud germination (15,17) are not traits with continuous variation, a contingency chi-square (X2) test (14) was used to analyze associations between these traits and smut infection percentages of R, MS, and HS cultivars.

Bud germination type, time and shoot growth rates

During October 1986 (Experiment I), 15 two-bud cuttings were obtained from the middle sections of stalks of each clone, and the upper bud of each cutting was excised. All cuttings of each clone were then dipped in a mixed fungicide suspension containing 0.1 g a.i./liter benomyl (0.2 g 50% wettable powder) and 0.7 g a.i./liter captan (1.4 g 50% wettable powder) for 20 minutes and then placed in a plastic box containing wet paper towels which served as a germmation chamber. Boxes were covered with black plastic and incubated at 25 C. The time required for bud germination, the type of germination, and shoot growth rate were recorded for each clone. Bud germination time was the number of days required for the first emergence of a shoot initial from the bud. Bud germination was classified as apical, subapical, or dorsal for individual buds (17) and compared among R, MS, and HS clones with a contingency chi-square analysis. Growth rate was calculated from the time visible signs of germination were observed until a shoot developed to a length of 10 cm. Correlations of bud germination time and shoot growth rate with smut infection percentages were determined for each clone. The experiment was repeated during November 1986 (Experiment II), except clone CP 76-340 was not included.

Shoot length-susceptibilitv relationship

During November 1986,135 and 170 one-bud cuttings were taken from the middle portions of stalks of two sugarcane cultivars, CP 65-357 and CP 74-383, respectively, and used to determine the effect of shoot length

7

on susceptibility to smut infection. All cuttings of each clone were dipped in a mixed fungicide suspension as described previously. They were then washed, dried and both ends were sealed with wax.

Thirty and 32 cuttings of CP 65-357 and CP 74-383, respectively, with ungerminated buds were dip-inoculated for 10 minutes in a smut spore suspension containing 5xl06 spores/ml. After being incubated in germination chambers at 30 C for 18 hours, they were washed with a 5% sodium hypochlorite solution to kill external smut spores and planted in 10-cm diameter clay pots in a sterile soil, sand, peat moss mix (3:l:l,v:v). The remaining cuttings of each cultivar were germinated in chambers covered with black plastic. Various numbers of germinated cuttings (Table 5) with shoot lengths ranging from 0.1-6, 6.1-12,12.1-18, and 18.1-24 cm were dip-inoculated, incubated, surface sterilized, and planted as described previously.

Primary shoots produced from inoculated cuttings of each clone were harvested such that four lateral buds were left on the cuttings in the pots. The growing point of each cutting was then stained with trypan blue to detect the presence of Ustilago scitaminea mycelium (13). Growing points were stained, mounted on glass slides, coverslipped, and examined under a compound microscope at 250X for the presence of fungal mycelium. If mycelium was not detected in the growing point of the primary shoot, then secondary shoots developing from lateral buds were examined. A plant developing from an inoculated cutting was regarded to be infected if mycelium was detected in any meristematic tissue.

Systemic infection evaluation

Fifteen clones (Table 6) were chosen for evaluation of resistance to systemic infection in individual plants. Six clones were previously rated as resistant in breeding program smut inoculation tests (unpublished), four clones were rated as moderately susceptible, and five clones were highly susceptible. Generally, 20 one-bud cuttings of each clone were selected from the middle portions of stalks and inoculated with smut spores using a wound-paste inoculation technique (9). Ungerminated buds were pin pricked with a needle to make six wounds and then painted with a smut spore paste. After inoculation, cuttings were incubated in black plastic bags containing wet paper towels for two days at room temperature and then planted in flats in sterile soil, sand, peat moss mix. When inoculated buds developed into shoots with three to four emerged leaves, they were transplanted into plots during June 1985 with distance intervals of 0.6 m between plants. Data recorded for each clone included the number of smut-free stools, the number of smut-infected stools, the total percentage of smut-infected stalks, the number of stools with resistance to systemic infection (stools containing apparently smut-free as well as infected stalks), the numbers of apparently healthy and smut-infected stalks in each stool, and the number of completely smutted stools.

RESULTS

Evaluation of association between bud traits and resistance to smut

In a smut inoculation test, seven clones were rated as resistant (R), four clones were rated as moderately susceptible (MS), and four clones were rated as highly susceptible (HS) (Table 1). Differences were detected between means for clones within and among resistance rating groups for bud length, width, and length x width (Table 1).

The major types of bud shape were round and ovate for R and MS clones and ovate for HS clones (Table 2). MS and HS clones did not have triangular buds, and obovate buds were not found for HS clones. R, MS, and HS clones had different bud shape frequencies (Table 2A).

The most frequent type of bud flange was medium for R, MS, and HS clones (Table 2B). The second most frequent type was small for HS clones and absent for MS clones. R, MS, and HS clones had different bud flange size frequencies. Absence of a bud groove followed by buds with shallow bud grooves had the highest frequencies for R, MS, and HS clones (Table 2C). MS clones did not have deep or very deep bud grooves. The percentage of buds with a deep bud groove was similar for R and HS clones, and only one bud with a very deep bud groove was found in a HS clone. Bud groove sizes varied among R, MS, and HS clones.

The most frequent type of bud germination was apical in R, MS, and HS clones in Exp. I (Table 3A). Dorsal and subapical types of bud germination also occurred in R, MS, and HS clones. In Exp. II, the most frequent type of bud germination was apical for R and HS clones and subapical for MS clones (Table 3B).

8

Dorsal bud germination was not observed. R, MS, and HS clones had different bud germination type frequencies in both experiments (Table 3).

Table 1. Comparisons of smut infection percentage means and resistance ratings with bud length, bud width, and bud length x width for 15 sugarcane clones.

Mean smut Bud length x

Sugarcane infection Smut Bud length (mm) Bud width (mm) width (mm2>

clone percentage rating1 Mean2 Range Mean Range Mean

CP 70-330 CP 72-356 CP 77-310 CP 67-412 CP 72-370 CP 66-346 CP 72-355

CP 65-357 L 65-69

CP 77-413 CP 73-308

CP 73-351 CP 77-407 CP 76-340 CP 74-383

0 0 0 1 1 2 5

12 17 21 21

31 37 43 57

R R R R R R R

MS MS MS MS

HS HS HS HS

6.8 6.0 7.1 5.8 8.5 9.7

12.2

5.7 6.3 8.1 7.3

6.8 10.5 7.7 7.4

6-8 4-7 5-9 5-7 6-12 8-12 8-18

4-7 6-7 6-10 5-10

4-8 7-15 7-10 5-9

6.0 6.2 6.6 6.6 7.8 6.9 8.9

5.8 5.5 7.1 7.0

5.5 9.5 7.4 5.7

4-7 4-7 5-9 5-8 6-11 6-9 6-13

4-7 5-7 6-8 5-10

4-8 7-13 6-8 5-8

40.8 38.0 47.6 38.5 68.0 67.7

111.0

33.3 34.8 57.8 52.0

38.0 101.2 57.4 42.3

LSD0.05= 0.55 0.44 7.53

1 R = resistant, MS = moderate susceptible, and HS = highly susceptible smut reactions.

2 Mean values were based on 40 measurements. Means within a column were analyzed by Fisher's Protected LSD.

Differences in the mean number of days required for bud germination were detected between clones within and among different resistance rating groups in both experiments (Table 4). In addition, the means for number of days required for bud germination were different between Experiment I and Experiment II for all seven R clones, two of four MS clones, and one of four HS clones. There was no consistent pattern in the changes in times required for bud germination between experiments.

Differences in the mean growth rate (mm/day) were detected between clone means within and in different resistance rating groups in both experiments (Table 4). The growth rate decreased for 14 clones in Experiment II, and the decrease was significant in four of seven, two of four, and two of three R, MS, and HS clones, respectively (Table 4).

Correlation coefficients between clone smut infection percentages and bud length (0.08), bud width (-0.01), and bud length x width (0.04), were all nonsignificant. In Experiments I and II, clone smut infection percentages were also not significantly correlated with the time required for bud germination (r = -0.25 and 0.08, respectively) or initial shoot growth rates (r = -0.11 and -0.09, respectively).

9

Table 2. Contingency chi-square analysis of associations between smut reactions and A. bud shape, B. bud flange, and C. bud groove in 15 sugarcane clones.

A. Bud s]

Smut reactions of tested clones1

R MS HS

B. Bud flange

Smut reactions of tested clones

R MS HS

shape

Round

44.0 53.3 10.0

ange

C. Bud groove

Smut reactions of tested clones

R MS HS

Percentage of buds with each bud shape tvpe

Oval Obovate Ovate

2.5 6.7 36.6 4.9 2.5 39.3 13.1 0.0 76.9

Percentage of buds with each bud flange tvpe

Absent Small Medium

1.1 37.3 47.5 31.2 13.1 54.1 5.0 25.6 60.0

Percentage of buds with each bud groove tvpe

Triangular

10.2 0.0 0.0

Large

14.1 1.6 9.4

Absent Shallow Deep Very deep

50.7 31.0 18.3 74.6 25.4 0.0 55.0 22.5 21.9

0.0 0.0 0.6

Chi-square value2

Value (X2) Probability

142.6 < 0.001

Chi-square value

Value (X2) Probability

127.2 < 0.001

Chi-square value

Value (X2) Probability

38.9 < 0.001

Seven clones were resistant, four were moderately susceptible, and four were highly susceptible to smut. R=resistant, MS = moderately susceptible, and HS = highly susceptible smut reactions.

Chi-square values calculated from actual numbers with each bud character.

Shoot length-susceptibility relationship

High smut infection percentages resulted in plants of both clones inoculated with initial shoots up to 6 cm long, whereas plants inoculated as ungerminated, intact buds developed low infection percentages (Table 5). Infection percentages then decreased to low levels in plants inoculated with initial shoots 6.1-24 cm in length (Table 5).

10

Table 3. Contingency chi-square analysis of association between type of bud germination and smut reactions of 15 sugarcane clones in experiments conducted during I. October and II. November, 1986.

A. Experiment I.

Smut reactions of tested clones1

R MS HS

B. Experiment II.

Smut reactions of tested clones

R MS HS

e Percentage of buds with

each bud germination tvoe Dorsal Subapical

24.5 23.3 5.4

16.7 11.6 21.4

Apical

58.8 65.1 73.2

Percentage of buds with each bud germination tvoe

Dorsal Subapical

0.0 0.0 0.0

48.4 69.8 22.0

Apical

51.6 30.2 78.0

Chi-square value2

Value (X2) Probability

10.1 <0.05

Chi-square value

Value (X2) Probability

19.3 < 0.001

1 Seven clones were resistant, four were moderately susceptible, and four were highly susceptible to smut. R = resistant, MS = moderately susceptible, and HS = highly susceptible smut reactions.

2 Chi-square values calculated from actual numbers with each bud character.

Evaluation of resistance to systemic infection

Smut infections developed in plants of two of six R clones, four of four MS clones, and five of five HS clones following wound inoculation (Table 6). The proportion of the total number of stools showing a smut infection was 11% and 20% for the two R clones compared to total percent stalk infection levels of 2% and 6%, respectively (Table 6). The infection levels for the MS clones ranged from 35-100% and 26-100% for stools and stalks, respectively, and infection levels for the HS clones ranged from 78-100% and 66-97% for stools and stalks, respectively. The proportion of the smut-infected stools of each clone that contained at least one apparently smut-free stalk (stools with resistance to systemic infection) was 100% for the two R clones and ranged from 0-71% for both MS and HS clones (Table 6). The mean percentage of infected stalks per stool with resistance to systemic infection ranged from 15-29%, 75-92%, and 60-93% for R, MS, and HS clones, respectively (Table 6).

DISCUSSION

None of the traits evaluated in this study including bud length, bud width, bud length x width, bud shape, bud groove, bud flange, time required for bud germination, type of bud germination, and initial shoot growth rate were consistently associated with smut resistance. These results indicate that none of these traits, as measured, can be used to reliably identify resistant or susceptible clones.

11

Table 4. Comparisons of smut infection percentage means and resistance ratings with length of time required for bud germination and initial shoot growth rates for 15 sugarcane clones in experiments conducted during I. October and II. November, 1986.

Mean Time (days) required smut for bud germination2 Mean growth rate3

Sugarcane infection Smut Experiment Experiment clone percentage rating1 I II I II

CP 70-330 CP 72-356 CP 77-310 CP 67-412 CP 72-370 CP 66-346 CP 72-355

CP 65-357 L 65-69

CP 77-413 CP 73-308

CP 73-351 CP 77-407 CP 76-340 CP 74-383

0 0 0 1 1 2 5

12 17 21 21

31 37 43 57

R R R R R R R

MS MS MS MS

HS HS HS HS

4.94

4.4 1.9 4.6 2.3 2.1 2.6

5.1 4.1 2.0 2.9

4.2 3.0 1.7 1.8

2.6**5

2.8** 3.1** 2.4** 3.6** 3.1* 4.0**

3.5* 3.6 3.1** 3.4

3.4 4.1

- 6 3.0**

9.6 9.8

10.4 9.4

12.2 10.2 13.1

10.3 9.5

10.0 12.1

10.1 10.6 9.8

13.6

8.9 9.6 8.2

10.0 9.2** 7.8** 8.1**

9.8 7.4* 9.5 7.7**

8.9 7.7**

8.5**

LSD0.05 = 1.05 0.82 1.44 1.23 1 R = Resistant, MS=moderately susceptible, and HS = highly susceptible smut reactions.

2 Mean of time (days) required for bud germination for 15 buds of each clone. 3 Means for initial shoot growth rates (mm/day) from germinated buds of each clone were determined from

the time (days) required for shoot lengths to reach 100 mm.

4 Clone means were generally based on 15 measurements. Means within experiments were analyzed by Fisher's Protected LSD.

5 Means for tested clones in Experiment I and Experiment II were compared in a t-test and some differed significantly at the P<0.05 (*) or P<0.01 (**) levels.

6 Clone CP 76-340 was not available in Experiment II.

Methods used to estimate bud size have varied between studies. Waller (17) used the amount of water displaced by excised buds, and the correlation between smut incidence and increasing bud size was 0.895. Muthusamy (11) reported bud area in cm2 and a correlation coefficient with smut incidence of 0.553. However, the method for calculation of bud area was not stated. Despite the differences in measurement methods used, the results of this study indicate that small bud size is not associated with smut resistance.

Correlation coefficients of 0.762, 0.796, and 0.768 were found between smut incidence and bud germination type, time to bud burst, and growth rate, respectively, in 18 clones in the study conducted by Waller (17). He concluded that dorsal germination and slow germination and initial growth rates were associated with resistance. Muthusamy (11) suggested that subapical germination was associated with resistance. In this study,

12

these traits were not consistently associated with smut resistance in either of two experiments. The frequency of round buds and dorsal or subapical germination was higher in R and MS clones, whereas HS clones showed a strong tendency towards ovate buds and apical germination. However, there was enough variation among clones within the R and MS groups to make these traits unreliable for prediction of resistance of susceptibility. In addition, the types of bud germination, the times (days) required for bud germination, and initial growth rates of R, MS, and HS clones differed significantly between Experiment I and Experiment II, conducted during October and November of 1986.

Table 5. Numbers of smut infections resulting in plants moculated as ungerminated buds or with primary shoots of increasing length in two sugarcane clones, CP 65-357 and CP 74-383.

Shoot length interval (cm)1

0 0.1-6 6.1-12 12.1-18 18.1-24

No. of buds inoculated2

CP 65-357

30 34 30 26 15

CP 74-383

32 42 36 40 20

No. of infected plants

CP 65-357

0(0) 7(0) 2(1) 1(0) 1(1)

CP 74-383

2(0) 20(0) 1(0) 3(0) 1(0)

Smut infection

percentage CP 65-357 CP 74-383

0 6 21 48 7 3 4 7 7 5

Single-bud cuttings of each clone were inoculated with smut spores as ungerminated (0) or germinated with primary shoot lengths categorized into intervals of 0-6, 6.1-12, 12.1-18, or 18.1-24 cm.

Cuttings were dip-inoculated in a smut spore suspension for 10 minutes and incubated in a germination chamber at 30 C for 18 hours, surfaced sterilized, and planted.

Smut infections were determined by the observation of fungal mycelium in apical meristems of the primary shoots or lateral buds. The number in parentheses indicates the portion of smut-infected plants in which the infection was detected in the secondary shoots but not in the primary shoot.

Previous studies suggested that environmental conditions can affect morphological and growth characteristics (8,11,17). In this study, it appeared that environmental conditions prior to cutting, most likely temperature, affected bud germination times and initial shoot growth rates. Growth rates for many clones were significantly lower for cuttings obtained during November; however, no consistent pattern was evident in the changes in times required for bud germination.

An interaction between growth rate and the length at which developing shoots of a clone become resistant to infection affects the period of time a shoot is susceptible to infection and the chance of coming into contact with smut spores. The results of the experiment to determine the length at which initial shoots of two cultivars, CP 65-357 and CP 74-383, became resistant to infection were very similar to the results of a study conducted with one cultivar by Bock in Kenya (3). In both studies, susceptibility decreased sharply after shoot lengths exceeded 5-6 cm. Infections developed at low frequencies in plants inoculated at shoot lengths ranging from 6-20 cm. No infections developed in plants inoculated at shoot lengths greater than 20 cm in Kenya, but infections did occur at low frequencies in this study in plants of both cultivars inoculated with shoot lengths ranging from 18-24 cm. In one study conducted in India (11), bud sprouting associated with insect damage was correlated with high infection levels in 10 susceptible clones. In another study (12), sprouted buds of one of two cultivars were susceptible to infection. However, no information was given concerning shoot lengths.

A disagreement exists concerning whether or not smut spores can infect ungerminated sugarcane buds on standing cane (3,17). When buds were exposed to smut spores for 18 hours prior to germination, six of 32 (18.8%) of the CP 74-383 plants became infected, whereas none of 30 CP 65-357 plants developed an infection. The results suggest that the level of susceptibility of ungerminated buds to infection varies among clones and may

13

be related to clone susceptibility. In addition, the frequency of infection in clones susceptible to infection in an ungerminated state appears to be low.

Table 6. Characteristics of infections resulting from wound inoculation with smut spores in resistant (R), moderately susceptible (MS), and highly susceptible (HS) sugarcane clones.

Clone

CP 61-37 CP 67-412 CP 70-321 CP 72-356 CP 72-370 CP 76-301 CP 65-357 CP 74-383 CP 78-303 CP 78-304 CP 80-306 CP 80-319 L 80-38 L 80-45 L 81-8

Smut rating

R R R R R R

MS MS MS MS HS HS HS HS HS

1 Smut infected sugarcane

Total number of stools

20 20 20 17 18 18 17 18 20 10 18 19 17 16 17

No. of smut-free stools

20 16 20 17 18 16 4 4

13 0 4 1 2 2 0

stools showing at least

No. of stools with

tance to syste­mic infec­tion1

0 4 0 0 0 2 9 1 5 0

10 1 9 0

10

No. of comple­tely infec­ted stools

0 0 0 0 0 0 4

13 2

10 4

17 6

14 7

Infec­ted stools (%)

0 20 0 0 0

11 53 78 35

100 78 95 88 88

100

one apparently smut-free stalk.

Infec­ted stalks (%)

0 6 0 0 0 2

76 88 26

100 66 97 70 81 79

%of stools with resis­tance to systemic infection

0 100

0 0 0

100 69 7

71 0

71 6

60 0

59

%of infected stalks in stools with

to systemic infec­tion2

0 29 + 18

0 0 0

15 + 11 84+8 92

75 + 19 0

64 + 20 93

53 + 24 0

60 + 16

2 Value represents a mean except when data recorded only from a single stool.

Previous investigations have suggested that different types of resistance to sugarcane smut can be recognized (7,10). Type I (pre-infectional) and Type II (post-infectional) resistance were terms used in Hawaii to describe different patterns of disease expression in resistant clones (7). Clones with pre-infectional resistance had low incidence of infected plants but high disease intensity in individual plants, whereas clones showing post-infectional resistance had high disease incidence but low intensity. In South Africa (10), pre-infectional resistance was evaluated by measuring the concentration of glycosidic substances in bud scales which inhibited germination of smut spores, and post-infectional resistance was attributed to differences in the colonization rate of hyphae and frequency and type of haustorial development in infected tissues.

The resistance mechanisms which result in differences in disease expression among clones and the times during the infection process when they occur are not clearly understood. As a result, some confusion arises in the use of terms previously used to describe different forms of resistance. Terminology which describes the recognized patterns of disease expression would be resistance to infection and resistance to systemic infection.

Resistance to systemic infection apparently occurs in Louisiana sugarcane clones. Evaluation was difficult in the four resistant clones that did not develop infections. However, this type of resistance was probably active in these highly resistant clones since the barrier represented by the bud scales was pierced during inoculation. The evidence for resistance to systemic infection was strongest for the two resistant clones, CP 67-412 and CP 76-301, in which some plants became infected. In both clones, the proportion of infected stalks within infected stools was low. This resulted in overall low infection levels and resistant reactions for both clones. The absence of infection in CP 70-321 and CP 72-356 in this experiment is in contrast to results of another study (2) in which some smut-infected secondary shoots were observed in plants of both cultivars developing from buds inoculated with the outer bud scales removed.

14

The infection levels resulting from wound inoculation were high for three of the four MS clones and would have, by other standards, resulted in the assignment to them of a HS smut rating. These results suggest that the expression of resistance in these clones might be partially due to a barrier type of resistance. However, lower levels of resistance to systemic infection were detected in some MS and HS clones. The frequency of stools with resistance to systemic infection was low (6% and 7%) for two clones but higher (59-71%) for the other five clones, and the percentages of apparently smut-free stalks in infected stools ranged from 16-47%.

These results support studies conducted in Barbados (18), Florida (6), Hawaii (7), and South Africa (10) and indicate that mechanisms which limit the development and expression of disease in individual plants contribute to smut resistance.

Resistance to smut is complex and may be expressed in several ways. The relative importance of the different mechanisms of resistance is unclear. Differences may be detected among clones; however, the methods used and the results obtained have varied among studies (2,4,6,7,10,18). The evaluation of stalk infection percentage in clones in annual dip-inoculation tests over more than one season will effectively assess the overall smut resistance of a clone. Thus, the established method is apparently still the best method to reliably evaluate smut resistance in clones in a sugarcane breeding program.

ACKNOWLEDGEMENTS

The authors gratefully acknowledge the technical assistance of Lori B. Grelen with the field and laboratory aspects of this research. We also thank Dr. Arnold M. Saxton for his suggestions in analyzing data of this study.

REFERENCES

1. Antoine, R. 1961. Smut. Pages 326-354 in: Sugar-Cane Diseases of the World. Vol 1. J. P. Martin, E. V. Abbott, and C. G. Hughes, eds. Elsevier Publishing Co., Amsterdam. 542 pp.

2. Benda, G. T. A., and H. Koike. 1985. Sugarcane smut and the distribution of whips on uprights of sugarcane. Sugar Cane 4:15-17.

3. Bock, K. R. 1964. Studies on sugarcane smut (Ustilago scitaminea) in Kenya. Trans. Brit. Mycol. Soc. 47:403-417.

4. Byther, R. S., and G. W. Steiner. 1974. Comparison of inoculation techniques for smut disease testing in Hawaii. Proc. ISSCT 15:280-288.

5. Comstock, J. C, S. A. Ferreira, and T. L. Tew. 1983. Hawaii's approach to control sugarcane smut. Plant Dis. 67:452-457.

6. Dean, J. L. 1981. The effect of wounding and high pressure spray inoculation on the smut reaction of sugarcane clones. Phytopathology 72: 1023-1025.

7. Ferreira, S. A., J. C. Comstock, and K. K. Wu. 1980. Evaluating sugarcane smut resistance. Proc. ISSCT 18:1463-1475.

8. James, G. L. 1969. Smut susceptibility testing of sugarcane in Rhodesia. Proc. South Afr. Sugar Technol. Assoc. 43:85-92.

9. Leu, L. S., C. A. Wismer, J. Daniels, and P. B. Hutchinson. 1970. Co-operating programs for screening sugarcane varieties for resistance to disease (1). The Taiwan program. Sugarcane Pathol. Newsl. 4:36-37.

10. Lloyd, H. L., and M. Pillay. 1980. The development of an improved method for evaluating sugarcane for resistance to smut. Proc. South Afr. Sugar Technol. Assoc. 54:168-172.

15

11. Muthusamy, S. 1974. Varietal susceptibility to smut (Ustilago scitaminea Sydow) in relation to bud characters. Proc. ISSCT 15:289-291.

12. Singh, K., and T. R. Bhudraja. 1964. The role of bud scales as barriers against smut infection. Proc. 5th. All-India Conf. Sugarcane Res. and Devel. Workers, pp 687-691.

13. Sinha, O. K., K. Singh, and S. R. Misra. 1982. Stain technique for detection of smut hyphae in nodal buds of sugarcane. Plant Dis. 66:932-933.

14. Steel, R. G. D., and J. H. Torrie. 1980. Principles and Procedures of Statistics. 2nd ed. McGraw-Hill Inc. N. Y. 633 pp.

15. Van Dillewijn, C. 1952. Botany of Sugarcane. The Chronica Botanica Co. Waltham, Mass. 371 pp.

16. Waller, J. M. 1969. Sugarcane smut (Ustilago scitaminea) in Kenya. I. Epidemiology. Trans. Brit. Mycol. Soc. 52:139-151.

17. Waller, J. M. 1970. Sugarcane smut (Ustilago scitaminea) in Kenya. II. Infection and resistance. Trans. Brit. Mycol. Soc. 54:405-414.

18. Whittle, A. M., and D. I. T. Walker. 1982. Interpretation of sugarcane smut susceptibility trials. Tropical Pest Management 28:228-237.

16

THE PRODUCTION OF SEEDLINGS IN THE LOUISIANA, "L", SUGARCANE BREEDING PROGRAM1

Keith P. Bischoff, Joey P. Quebedeaux, and Fred A. Martin Department of Agronomy, Louisiana Agricultural Experiment Station,

Louisiana State University Agricultural Center, Baton Rouge, Louisiana

ABSTRACT

Each year the Louisiana, "L", sugarcane breeding program plants approximately 75,000 seedlings at the St. Gabriel Research Station. Over the past three decades, more economical and efficient means of producing and planting these seedlings have evolved. During winter, true sugarcane seed is germinated in flats containing a soil, sand and sphagnum moss mixture. Approximately three weeks after germination, the seedlings are transplanted to individual pots, then grown in the greenhouse until transplanting to the field in April.

Prior to 1960, the seedlings were potted into 7.5 cm clay pots containing sterilized soil. Transplanting of seedlings to the field was by hand. In 1960, clay pots were replaced with 5.7 cm peat pots. Because the peat pots were planted with the seedlings, a tractor drawn mechanical transplanter was used for the first time. In 1978, peat pots were replaced with Jiffy-7 peat pellets. Jiffy-7's did not require filling with sterilized soil; thus time and labor were saved. The Jiffy-7's were transplanted to the field using the same mechanical transplanter as the peat pots.

In 1982, the latest improvements in the production of sugarcane seedlings at LSU were implemented: styrofoam Todd planter flats and two high-speed, mechanical, seedling transplanters. The reusable flats are a handling unit. By planting seedlings of only one cross per flat, the possibility of misidentifying the pedigree of a particular seedling is reduced. Using two high-speed transplanters on a single draw bar, two rows are planted simultaneously, thus decreasing planting time. As a result of these improvements, more seedlings can be grown in less greenhouse space. The total cost and man-hour requirements to complete the seedling production phase of the breeding program has been decreased significantly.

INTRODUCTION

The primary objective of the Louisiana, "L", sugarcane variety improvement program is to efficiently develop improved sugarcane cultivars for the Louisiana sugarcane industry. Phases of the program include: crossing, seedling production, single stool selection, evaluation in line trials, replicated infield variety testing and replicated outfield variety testing (3).

In January of each year, approximately 150,000 viable seed from a total of approximately 200 biparental crosses are germinated under greenhouse conditions. From these, approximately 75,000 seedlings are planted in the field at the Louisiana Agricultural Experiment Station's St. Gabriel Research Station, St. Gabriel, Louisiana. Over a 13-year selection and testing regimen, it is possible that one or more of these seedlings may become a commercial cultivar.

Over the past three decades, the system of producing and transplanting these seedlings has changed. Today's system is more efficient and cost effective. The purpose of this paper is to describe the changes and improvements that have taken place during this time.

'Approved for publication by the Director of the Louisiana Agricultural Experiment Station as manuscript number 88-09-2433.

17

DISCUSSION

The process of germinating the true seed or "fuzz," as described by Breaux (3), has remained relatively unchanged over the past three decades. Handling of seedlings after germination through transplanting to the field, however, has undergone major changes.

Prior to 1960, seedlings were potted from germination trays into 7.5cm clay pots which were filled with a sterilized mixture of screened soil, sand and peat moss (6.5:1:3)2. After potting, the seedlings were placed on greenhouse benches and grown for 60 to 65 days. Young plants were then transported to the field in April where they were removed from the clay pots and planted by hand. Empty pots were gathered and cleaned for use in subsequent years. From 1960 through 1978,5.7 cm peat pots replaced the clay pots (1). In April, the plants were removed from the benches, placed in metal trays and transported to the field for planting using a mechanical New Holland transplanter (2).

There were several advantages of peat pots over clay pots. The extra work in gathering and cleaning clay pots for reuse was eliminated since the peat pots were planted along with the seedlings. Handling of the seedlings was easier since peat pots are not as heavy or fragile as clay pots. Peat pots utilized less greenhouse space than clay pots, and the mechanical transplanter could now be used.

From 1978 through 1981, Jiffy-7's3 were used in the seedling program. The Jiffy-7's were soaked in water until expanded and placed in metal trays. Seedlings were then transplanted into the Jiffy-7's, and taken to greenhouse benches and grown for 60-65 days following a similar routine as had been used with peat pots.

An advantage of using Jiffy-7's was the elimination of screening, mixing and sterilizing soil for filling pots. This resulted in a saving of time and labor during transplanting for greenhouse growth. The Jiffy-7's were lighter and used less greenhouse space than peat pots. The same New Holland mechanical planter was also used with Jiffy-7's when transplanting to field.

In 1981, the utilization of the Speedling system4 for the greenhouse growing of sugarcane seedlings was investigated. Due to the success of these experiments, the system was incorporated in the new construction of the sugarcane breeding complex at St. Gabriel.

The planting system included model 150 styrofoam Todd planter flats with 128,3.8 cm square cells filled with Jiffy-mix Plus5 potting media. With the Speedling system the entire flat is placed on a specially designed bench in the greenhouse which allows air pruning of the roots as the seedlings grow. This causes the seedlings to form vigorous, yet easily removable root balls in individual cells of the flats. The seedlings remain in the flats in the greenhouse for 60-65 days after which time the flats are transported to the field and the seedlings are removed and mechanically transplanted.

The Speedling system has several advantages over the former systems: no soil is screened, mixed and sterilized; the lighter styrofoam flats are easier to handle; decreased chance of accidental mixing of crosses; less labor intensive; less greenhouse space (Table 1); watering and fertilization is automated through the use of the traveling spray boom (Figure 1); and through the use of two Model 6000 high speed mechanical transplanters6

attached to a single drawbar (Figure 2), planting time in the field is decreased considerably (Table 2).

The amount of greenhouse space required to produce a seedling crop has decreased with each new method introduced. Whereas 183 seedlings per square meter could be grown using clay pots, 538 plants per

2Giamalva, M. 1986. Personal communication.

Manufactured by A/S Jiffy Products Ltd., Norway.

"Speedling Incorporated, P. O. Box 238, Sun City, FL 33586.

5Jiffy Products of America, 250 Town Road, West Chicago, IL 60185.

8Mechanical Transplanter Company, Box 1008B, Holland, MI 49423.

18

square meter are now produced using the styrofoam planter flats. This has allowed for expansion of the seedling program without greatly increasing resource use.

Table 1. The amount of greenhouse space utilized by each planting system as indicated by the number of plants per square meter of greenhouse area.

Planting system

Number of plants per square meter

Clay pots Peat pots Jiffy 7's Speedling

183 269 420 538

Figure 1. Watering and fertilization of seedlings with automated spray boom.

19

Figure 2. Transplanting seedlings to the field using two high-speed mechanical transplanters attached to a single drawbar.

Table 2. The rate of transplanting for each system as indicated by the number of plants placed in the field per day and the number of days taken to complete the planting operation.

Planting system

Number transplanted

Number planted per day

Days to complete transplanting

Clay pots Peat pots Jiffy 7's Speedling

60,000 65,000 65,000 75,000

3,000 7,500 7,500

25,000

20 9 9 3

20

It is important to transplant the seedlings to the field in a relatively short period of time. Transplanting rate has increased significantly over the years. Prior to 1960, approximately 60,000 seedlings were hand-planted at a rate of 3,000 per day. Today, through the use of the two-row Speedling planter, planting rate has been increased to about 25,000 seedlings per day, a more than eight-fold increase.

As a direct result of these improvements, the total cost and man-hour requirements to complete the seedling production phase of the breeding program has been significantly reduced.

NOTE: Mention of a specific vendor does not constitute an endorsement of that vendor to the exclusion of all others that may also be suitable.

REFERENCES

1. Anzalone, L., M. Giamalva and S. J. P. Chilton. 1960. Present method used for selecting sugarcane varieties at LSU, Report of the Department of Plant Pathology to the Contact Committee of the American Sugarcane League, pp. 5-12.

2. Anzalone, L., M. Giamalva and S. J. P. Chilton. 1965. Methods used to select disease resistant varieties of sugarcane at the Louisiana State University. Proc. ISSCT 12:1165-1173.

3. Breaux, R. D. 1972. Selecting commercial sugarcane varieties from large seedling and clonal populations. Proc. ASSCT 2(NS):60-61.

21

ECONOMIES OF SIZE IN THE LOUISIANA SUGARCANE PROCESSING INDUSTRY

Brian A. Chapman and Ralph D. Christy Research Associate and Associate Professor, respectively

Louisiana Agricultural Experiment Station Louisiana State University Agricultural Center

Baton Rouge, Louisiana

ABSTRACT

Structural changes in the Louisiana sugarcane processing industry prompted an investigation into the economic efficiencies being experienced as the industry adjust toward fewer and larger firms. Economics of scale estimates were sought as a measure of economic efficiencies. Several methods for estimating economies were discussed, and one was selected for analysis of the industry. Analysis of primary cross-sectional data from 1979 through 1985 using the statistical cost technique indicated that on average the industry is experiencing increasing returns to scale. Implications of the analysis pointed to the industry continuing to adjust toward few large firms.

INTRODUCTION

The sugarcane processing industry in Louisiana has been adjusting toward fewer and larger firms for more than a century. Ever since the industry started the transition from plantation-oriented milling to centralized commercial processing in the 1870's, the number of processors has generally declined, while the average tons of sugarcane processed per mill has generally increased. This latter adjustment has been in response to demands made on the surviving capacity by cane made available from exiting firms, and a general increase in the size of the cane crop. Although firms have grown larger over time, little is known about their relative efficiencies.

While many factors can affect structural changes in an industry, concentration within a food manufacturing industry, such as sugarcane processing, can be attributed in large part to cost economies enjoyed within certain ranges of output. This paper examines the relationship between processing cost and output for the sugarcane processing industry in Louisiana. Specifically, this paper seeks to determine whether the cost economies being experienced in the industry are positive, negative, or constant, and to identify the relative cost of output for various size categories of sugarcane processing firms

This paper proceeds by first describing structural changes within the Louisiana sugar industry over the past two decades. Second, the theoretical basis for the analysis is discussed, and prospects for the future of the Louisiana sugarcane industry are offered. Finally, the empirical results are reported. As fewer firms are available to process Louisiana's sugarcane crop, measures of the industry's economic performance will be of increasing interest to policy makers and industry participants.

Structural change in the Louisiana sugarcane processing industry

Changes taking place in the Louisiana sugarcane processing industry over the past several decades are reviewed by discussing four elements of market structure: 1) the number of raw sugar factories, 2) the grinding capacity per mill per day (size), 3) the average number of tons of sugarcane ground per mill per season, and 4) the average total operating cost for sugarcane processing. The number of factories operating in the state remained fairly stable from 1959 through 1973 (Figure 1). However, from 1973 through 1982 the number of mills dropped approximately fifty percent, averaging a decline of slightly more than two mills per year. Since 1982 the number has remained unchanged for an unprecedented six consecutive years.

The average per mill, per day, capacity in tons of cane ground has increased steadily from 1967 through 1985 (Figure 2). The average annual increase for the period was approximately 9.5 percent per year, and approximately 150 percent over the period.

22

The average tons of sugarcane processed per mill per season tended to increase over the period under examination (Figure 3). The percentage change between years is somewhat erratic, due to changes in the number of mills operating in the state in a given season, and the total tons of cane harvested in a given season. This latter variable tends to vary widely relative to the number of factories.

1967 1970 1975 1980 1985 Year

Figure 1. Number of sugarcane processing factories, Louisiana, 1967-85

1967 1970 1975 1980 1985 Year

Figure 2. Average tons of sugarcane ground per day per mill, Louisiana, 1967-85.

23

1967 1970 1975 1980 1985

Year

Figure 3. Average tons of sugarcane processed per mill per season, Louisiana, 1967-85.

In an effort to understand how processor operating cost behaved over time, two measures of cost were considered. The first of these, total cost, tended upward over time, but displayed dramatic up and down movements during certain periods, due in large part to increases and decreases in the cost of sugarcane (1). The cost of sugarcane is tied directly to the price received for raw sugar. In Louisiana, sugarcane growers typically receive 61 percent of the value of the raw sugar derived from the cane the grower ships to the processor. This arrangement accounts for the dramatic increase in total cost in 1973/74, when raw sugar prices reached historic high levels. Similarly, operating costs fell dramatically in 1975, when raw sugar prices returned to more normal levels.

A better understanding of the changes in operating costs as they relate to operating profits may be had by the second measure, total cost less the cost of cane, deflated to account for increases due to inflation (Figure 4). This latter measure indicates a sharp rise in cost from 1969 through 1975, and a general decrease in cost during subsequent years. The exception in this latter period is a sharp rise and fall in 1980 and 1981, respectively.

The general conclusions from the observations noted above are that the industry is adjusting toward fewer firms, and these firms are increasing their capacities. Further, the surviving firms are in fact processing increasing amounts of sugarcane, yet not always at decreasing cost per unit. This cost-output relationship is examined in the balance of this paper.

24

1967 1970 1975 1980 1985

Year

Figure 4. Average total sugarcane processing cost less cost of sugarcane, deflated for inflation, Louisiana, 1967-85.

The relationships between firm costs and output: theory and method

Microeconomic cost theory provides a basis for identifying the relationships existing between the cost and output of a firm. Of particular interest to an analysis of the cost economies existing in an industry is the set of theory related to average cost, or cost per unit of output. In cursory terms, the slope of an average cost curve yields the response of cost to changes in output. Cost theory is generally framed in two time horizons -short run cost theory and long run cost theory.

In the long run, all inputs are assumed to be proportionately and optimally adjustable for various levels of output. Firms are assumed to be able to select among various optimum firm sizes in terms of output. As firm size (output) increases, the firm can experience three possible returns to scales. The long run is applicable to time series analyses and situations which permit the rather unrealistic long run assumptions. For this particular analysis, neither is the case.

In the short run, the firm operates under some set of fixed inputs, hence firm size is fixed, and output is limited to a certain range of output. However, as in the case of the long run horizon, three possible cost economies may exist over the range of output - economies of size, diseconomies of size, and constant returns to size, which translate to decreasing cost per unit, increasing cost per unit, and constant cost per unit, respectively. The short run horizon applies to cross-sectional data like those used in this analysis, and can accommodate the assumptions required in this type analysis.

Short run cost theory proceeds from the production function to the cost function, the elasticity of which gives the response of cost to changes in output (3). With some enabling assumptions, the production function can be stated as a schedule of the maximum amount of output that can be produced from a specified set of variable and fixed inputs, given a state of technology. The general physical characteristics of a production

25

function are held by the Law of Diminishing Returns. The law insists that, given a set of fixed inputs, and successive equal increment of a variable input, total physical product, or total output will first increase at an increasing rate then increase at a decreasing rate to a maximum at which point the total output decreases. A production function can demonstrate all three of these relational forms. The sources of these relationships are generally said to result from the specialization and division of labor and technological factors. Neither source is easily or clearly identifiable without rigorous examination of the production process. Consequently, this analysis does not attempt to identify the sources of the economies estimated for the Louisiana sugarcane processing industry.

Production (processing) cost is a monetarized expression of the explicit and/or implicit inputs in the production process. Hence, a cost function by definition embodies a measure of the physical as well as the economic relationships between the fixed and variable inputs and output of the production function. Because of this definitional linkage between the production and cost functions, by the Law of Diminishing Returns are likewise imposed on the cost function. Furthermore, these relational forms dictate the cost economies experienced at various levels of output.

Alternative methods exist for measuring scale and/or size economies (4, 5). Engineering studies are based on technological or physical relationships between the capacity of a particular machine and its output. To the extent this approach applies to a single machine, it may lack broader economic implications.

The survivor test holds that competition within a market will drive out inefficient firms in the long run (4). Empirical estimation of economies of scale by this technique is performed by classifying firms within a market by size, and measuring the change in relative market share of these firms or their categories. The implications are that more efficient firms will increase market share. The limits of this approach are recognized when the nature of competition is known, and the resulting trade-offs between private efficiency and public efficiency are explored (6).

The statistical technique relates cost to output data to make inferences about scale economies (3). A major problem in applying this method is acquiring data, particularly from manufacturing firms where information is often regarded as too sensitive to share with public researchers. Because this caution occurred in only limited instances within the Louisiana sugarcane processing industry, the statistical technique could be employed to determine the existence of economies of scale for this industry.

A simple example of a total cost curve that satisfied the inverse curvature postulated in the production function is the cubic cost curve (2):

where are given parameters. The familar average cost and marginal cost associated with

the cubic cost curve is written as:

A convenient measure of economies of scale is given by the elasticity of cost, the elasticity of the cost curve with respect to output, written as:

26

where factor prices are assumed given. can be interpreted as:

economies of scale < = constant returns to scale if = 1

diseconomies of scales >

Using the cubic cost function and the elasticity of cost presented above, economies of scale can be measured for the Louisiana sugar mills.

EMPIRICAL RESULTS

Ordinary least squares regression was used to estimate the cost-output relationship within sugar processing industry in Louisiana. The functional form used to estimate the cost-output relationship is:

2 3 Cost = b + b Gnd + b Gnd + b Gnd

0 1 2 3 where

Cost = Total annual cost of processing sugar per plant.

Gnd = Total raw cane ground as a measure of output. 2

Gnd = Total raw cane ground squared. 3

Gnd = Total raw cane ground cubed.

Operating cost and return data, and selected physical data, representing fifteen Louisiana sugarcane processors for grinding seasons 1979 through 1985, were collected via personal interviews during the summer of 1986. These data were obtained from annual audited statements.

Table 1. Estimated cubic total cost function coefficients for Louisiana sugar mills for years 1979-85 and the average of years 1979-1985.1

Year Constant Gnd Gnd2 Gnd 3 R2 F Ratio

Term Bo B1 B2 B3

1979

1980

1981

1982

1983

1984

1985

Average 1979-85

24431592.83

18430069.26

1108681552

6819642.20

30818825.29

11930373.88

68379894.35

43827536.60

-335.7594 (3.068)

-2043996 (2.609)

-81.6395 (0.670)

-333309 (0.417)

-2843331 (2.193)

-1263464 (1.081)

-762.05 (2.688)

-456.6476 (2.725)

.0016 (3.184)

8.3366E-04 (2.915)

2.70062E-04 (0.737)

1.14184E-04 (0308)

933420E-O4 (2.289)

5.63153E-04 (1.309)

0.0029 (2.793)

1.6826 (2.884)

-2.2747E-09 (3.184)

-1.0117E-10 (3.065)

-23757E-10 (0.732)

-1.0076E-10 (0302)

-9.9269E-10 (2.300)

-7.1874E-10 (1.448)

-33418E-09 (2.854)

-0.0019 (2.942)

.800 14.698

.762 11.712

.347 1.948

.479 3.365

.669 7.408

348 4.442

.692 8.247

.735 10.161

1 Number in parenthesis are t-statistics.

27

The model was initially run using cross sectional data for each year. This approach yielded disappointing results, which were contrary to theoretical expectations (Table 1). The results yielded incorrect signs, yet significant coefficients, and reasonably high coefficients of variation. Efforts were made to improve the results of the remaining years by using log models, a quadratic functional form, changing dependent variables, and using other measures of firm output. These attempts failed to improve earlier results. Finally, running the model using a linear functional form yielded satisfactory results (Table 2).

Table 2. Estimated linear total cost function coefficients for Louisiana sugar mills for years 1979-85 and the average of years 1979-1985.1

Year

1979

1980

1981

1982

1983

1984

1985

Average 1979-85

Constant Term B0

440332.60

1583513.48

1720862.68

2192967.22

1810770.14

2204478.49

1418870.34

1086228.50

Gnd B1

11.0571 (2.4202)

9.0078 (2.6222)

7.9549 (3.2555)

7.4429 (2.2087)

9.4823 (2.5806)

8.2751 (3.5280)

9.4977 (3.0300)

7.8663 (2.2143)

R2

.6162

.4758

.3147

.4662

.5095

.2974

.4305

.5133

F Ratio

20.872

11.801

5.971

11.355

13.501

5.502

9.825

13.712

1 Number in parenthesis are t-statistics.

It was concluded that the data represented output ranges which were best described by a linear relationship between cost and output. Furthermore, it was suspected that averaging over the seven year period, would provide a more representative performance of the firm, because unique annual factors, beyond the control of management, were spread evenly across the seven years under examination. Therefore, an average of the output (Gnd) and total cost (Cost), deflated for inflation, for each firm over the seven year period was tested, and resulted in the expected statistical results. The results of this later cross-sectional model are provided below:

Cost = 1,086,228.5 + 7.866(Gnd) (2.124)

2 R = .51 F Ratio 13.712

Overall this model performed well in explaining the cost-output relationship in the Louisiana sugarcane processing industry during the period of study. The signs are as expected, and the R2 suggests this model fits

28

the data well enough to allow conclusions about cost economies in the industry. Moreover, coefficients for years other than 1981 and 1984 were significant at the 1% level. Coefficients for the exception years were significant at the 5% level. This model was chosen to compute the elasticity of cost, the chosen measure of economies of scale. Dividing average output by the average costs and multiplying by the coefficient B (Gnd) an cost elasticity of .6864 was obtained as shown below: 1

A coefficient of less than 1.0 suggests that on average the Louisiana sugarcane processing industry experienced increasing returns to scale during the period under study. If Louisiana sugarcane processors increase their average output, operating costs per unit should be expected to decrease.

CONCLUSIONS

As an industry becomes more concentrated, it becomes appropriate to examine the efficiency of firms operating in that industry. Significant structural changes have occurred in the Louisiana sugarcane processing industry. Over time, the industry has adjusted toward fewer and larger processors. To understand the economic implications of these structural changes, this study investigated size economies within the Louisiana sugarcane processing industry as a measure of economic efficiency. The empirical results of this analysis suggest that on average firms in the industry operated at increasing returns from 1979 through 1985. This suggest that average firm efficiency has improved as average firm size increased during this period.

The results of this study have implications for the future direction of the Louisiana sugar industry. They suggest further adjustments toward fewer and larger sugarcane processing factories are possible. As current and future technologies allow for increasing economic returns to factory size, firms can be expected to increase output in an attempt to reduce per unit cost and maintain or improve their competitive position. Firms unable to take advantage of greater outputs may find themselves in a weakening position. The process of adjustment and the rate at which it proceeds will be impacted in large part by the domestic demand for refined sugar and in the price received for raw sugar. A decrease in either should act to drive the industry toward further concentration.

REFERENCES

1. Campbell, Joe R. 1969-80, Raw Sugar Mills in Louisiana - Returns, Costs, and Profits for 1967 Grinding Season, DA.E. Research Report Nos. 395, 412, 427, 442, 474, 475, 493, 551, and 568, Department of Agricultural Economics and Agribusiness, Louisiana State University, Baton Rouge, Louisiana.

2. Intriligator, Michael D. 1978, Economentric Models, Techniques, and Applications. Englewood Cliffs, New Jersey: Prentice-Hall, Inc.

3. Johnson, J. 1960, Statistical Cost Analysis. New York: McGraw-Hill Book Company.

4. Koch, James V. 1980, Industrial Organization and Prices. Englewood Cliffs, New Jersey: Prentice-Hall, Inc.

5. Scherer F. M., 1980, Industrial Market Structure and Economic Performance. Boston: Houghton Mifflin. Boston.

6. Shepherd, William G. 1967, "What Does the Survivor Technique Show About Economies of Scale?" Southern Economic Journal, 34:113-22.

29

A METHOD FOR DETERMINING SPORE PRODUCTION OF SUGARCANE RUST, (Puccinia melanocephala).

J. M. Shine, Jr. Florida Sugar Cane League, Canal Point, Florida

Victor Chew USDA-ARS, Gainesville, Florida

and J. D. Miller

USDA Sugarcane Field Station, Canal Point, Florida

ABSTRACT

A method was developed for measuring spore production of sugarcane rust (caused by Puccinia melanocephala H. Syd. and P. Syd.) using a Coulter Counter1. Spore counts of suspensions of spores at concentrations higher than 5000 spores/ml did not differ significantly from counts made using a hemacytometer slide. The mean number of spores per ml and standard error using the Coulter Counter were 1247 and 102, and 1630 and 847 using the hemacytometer in a suspension containing approximately 1250 spores per ml. The time required to make determinations using the Coulter Counter was 33% less than the time required to make counts under the microscope. Leaves were collected from the field, sectioned, and spores were collected after an 18 hour incubation period. Accuracy of estimates of rust intensity were improved by sampling more than one field of the same clone and several rows in each field. The number of samples taken per row and number of readings taken per sample had little effect on accuracy. Large field areas may be evaluated efficiently with a relatively small number of samples. Mean rust spore production per unit leaf area for a given clone can be determined to a specified level of precision.

INTRODUCTION

Rust, caused by Puccinia melanocephala H. Syd. & P. Syd., is now an important disease of sugarcane. The disease was first discovered in Florida in 1979 by Dean et al. (2). Since that time numerous changes in host-cultivar resistance have been observed (1). A quantitative method for measuring disease severity is needed to evaluate sugarcane rust resistance both in agronomically acceptable clones intended for commercial production and in breeding lines to be used to develop cultivars with stable resistance to rust.

The most common method used in sugarcane to evaluate rust disease levels in Florida is a visual rating scale from 0-4 (8). This method is quite useful for rapidly screening large numbers of progeny and clones in a breeding and selection program. Methods for visually estimating percentage leaf area infected have been used to measure rust levels in both cereals and sugarcane (4,6). Neither method is quantitative nor sensitive enough to detect slight differences in disease intensity.

There is a positive correlation between the number of spores produced on infected wheat leaves and the progress of a rust epidemic (7). Methods to count the number of spores produced per unit area of leaf tissue have been employed in the rust-small grain pathosystems to evaluate varietal susceptibility to rust (3,4). These methods include counting collected spores produced per unit area of leaf tissue directly under the microscope using a hemacytometer (9) and weighing collected spores on an analytical balance (3). A Coulter Counter Model B was used in 1972 to assess intensity of yellow rust on European wheat cultivars (5). Quantitative differences in disease intensity were detected, but the method required extensive sample preparation.

This study presents an improved method using a Coulter Counter Model ZM (Coulter Electronics Limited, Northwell Drive, Luton, Beds., LU3 3RH England) to quantify the number of sugarcane rust spores

1Trade names are used in this publication solely for the purpose of providing specific information. Mention of a trade name does not constitute a guarantee or warranty of the product by USDA or any endorsement by the Department over other products not mentioned.

30

produced per unit area of leaf. The Coulter Counter Model ZM is designed to count particles of a known size suspended in an electrolytic medium containing a heterogeneous particle size distribution. The principle is based on the change in resistance of current as a particle passes through an aperture. This change in resistance is a function of the volume of particles. Calibration of the device with particles of known size allows the desired upper and lower particle size limits to be set.

The objectives of these experiments were to: 1) develop a method to remove all mature spores from leaves or leaf sections; 2) compare counts made with the Coulter Counter with the counts made with a Spenser Hemacytometer slide; 3) develop sampling methods for sugarcane leaves in the field to determine the necessary numbers of blocks or locations, rows, leaves and aliquots of spore suspension required to make efficient, accurate estimates of spore production per unit leaf area.

MATERIALS and METHODS

A Coulter Counter Model ZM calibrated for a 100 micron aperture tube was set to count particles within a range of 20.1m - 40.22m The spores collected typically ranged from 22-25m x 33-36m based on measurements made under a microscope. Though teliospores are known to occur in Florida none were observed on the samples collected in these experiments. The spores collected were observed to be urediospores. Spore counts made with the Coulter Counter were compared to counts made under the microscope with a Spenser Hemacytometer. The design of the slide and its calibration were described by Tuite (9).

Spores were collected from heavily infected-leaves of sugarcane cultivar CP 78-1247. Spores were collected by washing leaf sections taken from the top visible dewlap leaf in an ultrasonic bath (Mettler Electronics, Pasadena, Calif.). This concentrated spore suspension was diluted with distilled water to prepare three 250 ml suspensions containing approximately 1 x 105 spores/ml. The concentration of the three spore suspensions were estimated to be 17.9 x 104, 9.7 x 104, and 5.8 x 104, respectively. The suspensions were prepared to compare spore counts from serial dilutions between the Coulter Counter and the hemacytometer. The dilution was distributed as 1:99,1:9,1:1, and 1:0 parts original stock suspension to parts distilled water. Each dilution was prepared for 100 ml final volume. Each dilution was stirred constantly while nine, 5 ml aliquots were drawn and placed in 15 ml of Isoton II electrolyte. Three spore counts were made on each of the nine samples from the three stock suspensions using both methods. A second dilution series was conducted using a single stock suspension estimated to contain 1000 -1500 spores/ml. The spores were collected in electrolyte and the dilution series prepared and counted as described above.

A second experiment was conducted to devise a method to collect spores from leaves on cane growing in the field. Entire leaves were collected from each of two replicates of a field trial containing seven clones. The test was planted on August 26, 1987. Six rows 9-m-long and 1.5 m apart were planted in each plot in a randomized complete block design.

The top visible dewlap leaf was selected to evaluate rust intensity throughout this experiment. A total of 40 leaves were collected from each plot by removing the entire leaf blade from the plant. Two samples of five leaves were taken randomly from each of the four central rows in each plot from tillers of approximately the same height on March 15,1988. Leaves collected from each row were bundled in groups held by a rubber band at the base of the leaf and placed in a large plastic bag to prevent desiccation during transport.

Early investigations revealed that large amounts of debris adhered to field-collected leaves. This debris interfered with determinations made using the Coulter Counter. Spores adhering to the leaf and debris were removed by washing the leaf with distilled water delivered by a pressure sprayer operated at 60 psi. This was enough pressure to remove debris and moisten the leaf surface without fraying the leaves. The moist leaves were then placed basal end down in a bucket containing 4" of water and covered with a loose fitting plastic bag, then placed in a dark room at 25°C for 18 hours. One bucket contained four bundles of ten leaves representing each of the four rows sampled in each plot. The leaves were removed from incubation and samples prepared to count spores produced over the 18-hour-period.

Leaves were sectioned by measuring 30 cm from the tip of the leaf and taking ten, 1.25 x 2.5 cm rectangular sections at 5-cm-intervals in a basipetal direction on alternating sides of the mid-vein. The sections

31

32

were cut with a patch budder designed for budding pecan trees in nurseries. The cutting tool minimized leaf handling and provided a quick way to obtain sections of uniform size. The sections represented approximately 10 to 15% of the total surface area of each leaf.

Sections from the five leaves were placed in a 45 ml test tube containing 2 ml of Isoton II electrolyte solution. The tube was closed with a screw cap to prevent desiccation while preparing samples. An additional 28 ml of electrolyte was added to each test tube to collect spores from the leaf sections. This was sufficient to cover all of the tissue in the container. The test tubes containing the leaf sections in electrolyte were placed in a low frequency, low intensity ultrasonic cleaning bath for 20 minutes. Each tube was shaken vigorously to resuspend any spores that may have settled to the bottom of the tube before drawing the first aliquot. Three, 5 ml aliquots were drawn from each test, tube and each was placed in a cuvette containing 15 ml of electrolyte. Three spore counts were made on each of the prepared samples using the Coulter Counter. The estimated spore production per unit area of leaf surface based on the average of three determinations for the 18 hour incubation period was analyzed statistically using the model:

The calculated variance components were used to calculate the number of blocks or locations, rows, five-leaf samples and aliquots required to estimate the mean number of spores produced per unit area of leaf tissue on a given clone at a specified level of precision.

RESULTS and DISCUSSION

Spore counts made with the Coulter Counter Model ZM were compared to counts using a hemacytometer to test the accuracy of an unknown method against an accepted method of counting rust spores. The recommended usage of the hemacytometer requires 200 to 250 spores in one microscopic field of view to obtain an estimate within 10 - 15% of the true mean concentration of the stock suspension (10). Counts of this magnitude estimate concentrations on the order of 5 x 105 spores/ml of suspension. The same order of concentration of cells is recommended to obtain the most accurate results with the Coulter Counter. The first dilution series was designed to fit within the range of the highest degree of accuracy of both methods to compare their relative accuracies. Data presented in Table 1 show that the estimate of the spore concentration in the three-stock suspensions averages two times the expected result using the Coulter Counter and four times the expected result using the hemacytometer at the 1:99 concentration. The estimated mean number of spores in the stock suspensions at the 1:9,1:1 and 1:0 concentrations were not significantly different within each of the three suspensions regardless of the method, but the standard errors using the hemacytometer at the 1:9 and 1:1 concentrations were more than twice that of the Coulter Counter estimates. The most consistent results were obtained at the 1:1 and 1:0 concentrations using the Coulter Counter with the standard error averaging less than 2% of the estimated mean (Table 1).

The spore concentration in the 1:99 dilution from each suspension was less than 1000 spores per ml. Data from other experiments indicate the number of rust spores on leaves bearing fruiting pustules range from 100 - 2500 spores/cm. This would yield approximately 300 - 6000 spores/ml using our current sample preparation methods (unpublished data). A new suspension containing approximately 1250 spores/ml was made to determine whether concentrations in the range found on collected leaves could be accurately determined. The standard error of the 1:0 dilution was less than 10% of the estimated concentration using the Coulter Counter and more than 50% of the estimate using the hemacytometer slide (Table 2). The standard error at the 1:9 dilution using the Coulter Counter averaged near 10% of the estimate while the standard error using the hemacytometer averaged more than 40%. There was concern based on these data that neither counting method was sufficiently accurate to estimate the number of spores contained in dilute spore suspensions.

The estimate of the mean number of spores contained in the 1250 spores/ml stock suspension obtained with the Coulter Counter suggests that preparation of the dilution series was the largest contributer to error within the original 3 stock suspensions. The estimated mean of the 1250 spores/ml stock suspension calculated from the 1:99 dilution counts was more than 50 times greater than the expected mean with the hemacytometer and more than four times the expected mean with the Coulter Counter. The standard error was 27217 with the hemacytometer and 1599 using the Coulter Counter at the 1:99 dilution.

Table 1. Estimated mean number and standard error (SE) of rust spores per ml of sample prepared from three spore suspensions.

Expected Suspension 1 Suspension 2 Suspension 3 Conc.1 17.9xl04 9.7xl04 5.8xl04

Method Counter Slide Counter Slide Counter Slide

1 Expected concentration was calculated from the means of the estimated concentration of spores per ml in the parent suspensions at the 1:9, 1:1, 1:0 dilutions using both methods. The values obtained at each of these dilutions were not significantly different within each suspension.

2 Sample spores per ml means and standard errors were calculated from three spore counts per sample on each of nine samples.

3 Parts of stock suspension to distilled water. Five-ml samples from each dilution were placed in 15 mis of electrolyte, therefore the counts presented are multiplied by a factor of 4 times the dilution factor to obtain spores per ml of each stock suspension.

Results obtained at the 1:1 and 1:0 concentrations with the Coulter Counter were similar, the standard error averaging less than 10% of the starting concentration (Table 2). This level of accuracy was not expected since this implies that suspensions containing as few as 500 spores/ml, possibly, could be counted to within 10 - 15% of the true mean population. This degree of accuracy has not been reported with the hemacytometer. The estimated number of spores in the 1250 spores/ml suspension at the 1:0 concentration with the hemacytometer was close to the expected result, but the standard error was more than 50% of the estimated mean.

33

Table 2. Estimated mean number and standard error (SE) of rust spores per mil of sample prepared from one stock suspension containing approximately 1250 spores per ml.

1 Concentration in parts of stock suspension to Isoton II electrolyte.

Data presented here for counting dilute suspensions of rust spores are in agreement with similar data presented on bloodwork by the manufacturer. The most important difference between counting blood components and spores produced by fungal pathogens of plants is the fact that blood is essentially sterile and contains no foreign particles. Furthermore, blood components are suspended in a water soluble medium. Spores of many species of fungi, including P. melanocephala, are strongly hydrophobic and are extremely difficult to suspend uniformly in water without adding surfactants. The spores tend to adhere to the leaf after release from the pustule and spores being released from the pustules tend to clump, particularly under humid field conditions. These characteristics make working with sugarcane rust more difficult than rusts of cereals which produce large numbers of dry spores readily collected by tapping the leaf over a vessel. We were aware at the outset of these experiments that this might be an important barrier to overcome.

Extraneous spores and leaf debris must be minimized to obtain accurate counts with the Coulter Counter. Several cleaning methods were attempted. The most reasonable method appeared to be washing the leaves with

a pressure sprayer using a small amount of water. Leaves had to be incubated for a period of time for fresh spores to be produced after washing. A test was designed to determine the feasibility of washing with the pressure sprayer and to evaluate incubation periods and conditions to optimize spore production and enumeration. Spore counts from washed leaves were approximately 10% of counts from unwashed leaves when leaves were collected from the field and prepared immediately for counting. Other leaves were held in covered buckets for 18 and 36 hours. Spore production was measurable after 18 hours and differences in rust intensity could be detected. Visual observations of washed leaves at 24 hours indicated that many spores had germinated, and mycelia from rust and secondary fungi had begun to cover the leaves. Samples prepared from leaves held for 36 hours after washing were difficult to analyze because of mycelial growth present, and the resulting counts were highly variable.

A procedure to collect and handle leaves from which spore production was to be measured was developed to minimize sample variation. After observing a wide range of rust infection, we chose the top visible dewlap leaf to best represent rust spore production on the majority of clones. Initial infection of sugarcane leaves primarily occurs while leaves are emerging from the whorl. The disease on the top visible dewlap leaf has typically completed its primary infection cycle, pustules are reaching their peak spore production potential and

34

pustules formed from secondary infections are generally absent. Older leaves are often necrotic and infected with secondary pathogenic and saprophytic fungi.

The leaf sectioning procedure evolved after attempting a variety of methods. Initially, leaves were cut into sections 5-cm-long beginning 20 cm back from the leaf tip. It was noted that the number of pustules occurring on the leaf decreased basipetally, but no numerical relationship was derived. We decided that three, 5-cm sections representing 15 - 20% of the total leaf surface area provided an adequate sample. Sectioning the leaves in this manner was time consuming since the area of each section had to be measured. The patch budder, used to obtain sections of uniform size, greatly decreased the time required to cut the leaf sections. Single-leaf samples were used initially; 10 - 30, single-leaf samples were collected from plots. The coefficient variation using single-leaf samples was greater than 50%, too high to obtain acceptable accuracy in results. Combining sections from five leaves eliminated this wide degree of variation encountered with single-leaf samples.

The number of five-leaf samples required to detect differences in rust spore production among clones needed to be determined before implementing this counting technique to evaluate rust intensities in experiments. The expected mean squares were used to calculate the variance components listed in Table 3 for locations or blocks, rows, five-leaf samples, and aliquots from each sample.The variance components were used to determine the numbers of each sampling unit required for a desired level of accuracy. A coefficient of variation (CV) of 20% was accepted as a suitable threshold for accuracy. The overall mean square root of the number of spores produced per square cm of leaf was 14.25 and the maximum acceptable value of the standard error was 2.85 for a CV of 20%. The corresponding variance was 8.12.

Table 3. Analysis of variance of the square root of spores per square centimeter of abaxial leaf surface area.

Source Degrees of freedom Mean square Variance components

Clones 6 4,458 Blocks 1 600 1.820 Clones x blocks 6 295 9.050 Rows (clones x blocks) 42 78 11.960 5-leaf samples (r x b x c) 56 6 1.178 Aliquots (s x r x b x c) 224 3 2.906 Total 335 Grand mean = 14.250

Values of 1 and 5 were used for the levels of the sampling variables in Table 4 to illustrate effects of high and low levels of each sampling variable combination on the experimental CV. The table was constructed as a guide for designing experiments to record rust disease progress on clones representing a wide range of rust intensity.

Increasing the number of rows sampled in a large planting from 1 - 5 decreases the CV by 10% when only one field is sampled. Increase in the number of sampled fields from 1 - 5 decreases the CV by 12 to 20% (Table 4). The number of five-leaf samples and aliquots prepared have little effect on the reduction of the CV. An estimate of rust intensity over a larger area could be obtained by sampling one row in five or more fields. Fields can be sampled randomly to obtain disease progress curves for different clones over time to study many aspects of the pathogen and the disease.

Developing a technique to quantitatively measure sugarcane rust intensity has become an important need with the increased presence of the disease in the Everglades Agricultural Area since 1979. The absence of freezing temperatures in the winters of 1986-1987 and 1987-1988 may have lead to the recurrence of epiphytotics. Several edaphic factors have been observed to influence disease intensity and clonal reactions to the disease vary over time (R. Raid, personal communication). There is a concensus that rust races are involved in the south Florida area epidemics, but the extent is entirely conjectural.

35

Table 4. Estimated coefficient of variation (CV) with changing numbers of sampling variables.

We hope that the reported method of evaluating spore production of sugarcane rust will aid investigators in examining all aspects of this disease that is becoming more of a concern to Florida sugarcane growers.

ACKNOWLEDGMENTS

The assistance of the Palm Beach County School Board Executive Intern Program through students David Honeycutt and James Underwood is appreciated. The assistance of Wayne Jarriel is acknowledged, also.

REFERENCES

1. Dean, J. L. and L. H. Purdy. 1984. Races of the sugarcane rust fungus Puccinia melanocephala, found in Florida. Sugarcane 1:15-16.

2. Dean, J. L., P.Y.P Tai and E. H. Todd. 1979. Sugarcane rust in Florida. Sugar Journal 42(2): 10.

3. Johnson, R. and D. E. Bowyer. 1974. A rapid method for measuring rust production of yellow rust spores on single seedlings to assess differential reactions of wheat cultivars with Puccinia striiformis Ann. Appl. Biology 77: 251-258.

4. Peterson, R. F., A. B. Campbell, and A. E. Hannah. 1948. A diagrammatic scale for estimating rust intensity on leaves and stems of cereals. Canadian Journal of Research 26: 496-500.

5. Preistley, R. H., and D. A. Doling. 1972. A technique for measuring the spore production of yellow rust Puccinia striiformis on wheat varieties. Proceedings of the European and Mediterranean Cereal Rusts Conference, Prague 1972, I, 219-223.

6. Purdy, L. H. and J. L. Dean. 1981. A system for recording data about the sugarcane rust/host interaction. Sugarcane Path. Newsletter 27: 35-40.

36

7. Schafer, J. F. and A. P. Roelfs. 1985. Estimated relation between numbers of urediniospores of Puccinia graminis f. sp. 'tritici' and rates of occurrence of virulence. Phytopathology 75: 749-750.

8. Tai, Peter Y. P., Jack L. Dean, and J. D. Miller. 1979. Frequency of rust susceptibility in the sugarcane variety development program at Canal Point. Proc. ASSCT 9(NS):40-43.

9. Tuite, John. 1969. Plant Pathological Methods Fungi and Bacteria. Burgess Publishing Co., Minneapolis, MN. 239 pp.

10. Wilson, P. W. and S. G. Knight. 1952. Experiments in Bacterial Physiology. Burgess Publ. Co., Minneapolis.

37

CROP-HERBICIDE MANAGEMENT OPTIONS FOR JOHNSONGRASS CONTROL IN FALLOWED SUGARCANE FIELDS1

Edward P. Richard, Jr. Research Agronomist, Sugarcane Research Unit, ARS, USDA,

Houma, Louisiana 70361 and

Howard P. Viator Resident Director and Professor, Louisiana Agricultural Experiment Station

Jeanerette, Louisiana 70544

ABSTRACT

Three crop-herbicide management options were imposed on fallowed sugarcane fields in 1981, 1982 and 1984. Crop options included fallow only, wheat + fallow, and fallow + soybeans. Within each option, three levels of johnsongrass control were imposed by the use of plowing or metribuzin at 0.4 kg ai/ha (low/none), trifluralin + metribuzin at 1.1 + 0.4 kg/ha (medium), and trifluralin + metribuzin at 12 + 0.4 kg/ha followed by two postemergence applications of sethoxydim at 0.56 kg/ha (high). In the fallow-only system, the high level of herbicide usage was more effective than the medium level in preventing johnsongrass development. The high level was also more effective than plowing only where plowing was limited to June and July (2 yrs) but not where an additional plowing was performed in August (1 yr). Planting wheat in the winter, following the destruction of the second ratoon, did not decrease total weed cover and johnsongrass foliar cover and panicle production. Johnsongrass control in the option involving early fallow disking and double-drilling of soybeans on reformed 1.8 m sugarcane rows treated with trifluralin + metribuzin at 1.1 + 0.4 kg/ha (medium level) was equivalent to fallowed plots receiving the high level of herbicide usage. The results indicate that herbicide usage is essential to insure consistency in a johnsongrass control program for fallowed sugarcane fields and that the growing of soybeans during the fallow period provides an additional increment of control over herbicide treatment alone.

INTRODUCTION

In Louisiana, sugarcane (Saccharum interspecific hybrids) is routinely grown as a 3-yr crop with annual harvests on raised beds spaced 1.8 m apart. Cultural practices conducive to successful sugarcane growth are also favorable for growth of perennial johnsongrass [Sorghum halepense (L.) Pers.], a major weed of sugarcane in Louisiana. Because selective herbicides provide only partial control of rhizome johnsongrass in sugarcane, the potential for yield suppression from johnsongrass competition increases with each crop season with the severity of this competition, often reducing the crop's longevity.

Fields are destroyed by disking in late fall or spring following the harvest of the last ratoon crop and remain fallow during the spring and summer months. Disking may be repeated six or more times during the fallow period to destroy johnsongrass rhizomes which may have developed during the 3-yr crop cycle and to deplete soil stores of weed seeds. Success of this program is weather-related with rainfall during the summer months frequently preventing these timely diskings. This may allow newly-emerging johnsongrass seedlings to replenish the soil reserves of seed and rhizomes depleted earlier.

Chemical-fallow programs employing the use of preemergence herbicides have been developed for several crops to minimize erosion and/or to reduce soil moisture loss associated with frequent diskings (1, 2, 3). In Louisiana, metribuzin [4-amino-6-(l,l-dimethylethyl)-3-(methylthio)-l,2,4-triazin-5(4H)-one] has been shown to provide 8 to 10 weeks of seedling johnsongrass control in fallowed sugarcane fields that were previously disked at frequent intervals to destroy existing rhizomes (5). Obtaining preemergence herbicide persistence for only a portion of the 5- to 6-month fallow period without the use of postemergence herbicides to destroy escaped weeds requires high rates in an area of the field where no short-term economic gain can be anticipated. This study was initiated to investigate the feasibility of using annual crops during the fallow period to suppress johnsongrass growth and, thereby, reduce herbicide rates, and to provide additional farm income.

1 Research was part of Southern Regional Project S-159.

38

MATERIALS AND METHODS

Fields of second ratoon sugarcane growing on a Mhoon soil (fine-silty, mixed nonacid, thermic Typic Fluvaquents) and severely infested with rhizome johnsongrass were selected. Immediately after sugarcane harvest on December 9, 1981 (study A), October 25, 1982 (study B), and December 12, 1984 (study C), the sugarcane ratoon, associated plant residues, and the existing beds were destroyed by four successive passes with a disk harrow. A level seedbed was formed using two passes with a S-tined seedbed conditioner prior to drill-seeding wheat (Triticum aestivum L.) at a 21 cm drill spacing on designated plots on December 12 ,1981, (study A), December 23,1982 (study B), and December 18, 1984 (study C). Because of a poor stand in 1984, the entire field was redisked once and a new seedbed prepared using one pass with the S-tined seedbed conditioner on March 12,1985. Designated wheat plots were reseeded on March 13, 1985. Since little was known about wheat performance in southeast Louisiana, several cultivars were evaluated. In studies A and B, wheat plots were divided in half lengthwise and cultivars 'Coker 65-15' and 'Florida 301' planted in study A and cultivars 'Coker 762' and 'Coker 797 planted in study B. For study C, the entire plot was seeded to 'Coker 747'.

Conventional 1.8 m wide by 35 cm high sugarcane rows were reformed on the plots not planted to wheat on March 16, 1982 (study A), March 26, 1983 (study B), and April 11, 1985 (study C). Row buildup was accomplished using a plow to mark wheel furrows followed by two passes with a four-disk per gang rolling bed chopper to build up the rows. Weed vegetation was not destroyed prior to row formation. Soybeans [Glycine max (L.) Merr. (cultivar 'Forrest')] were seeded on May 6, May 27, and May 3, in two drills spaced 61 cm apart on top of each 1.8 m bed in studies A, B, and C, respectively. Prior to planting soybean seeds were coated with a commercial mix of Rhizobium bacterium. To meet the fertility needs of the young plants in the interim between inoculation and symbiotic nitrogen fixation by the nodulated bacteria, ammonium nitrate (33-0-0) was banded between the two drills of soybeans at 11 kg N/ha when soybeans were in the two to three leaf stage.

Within each cropping system, three herbicide levels (none or low, medium, and high) were imposed to produce various levels of johnsongrass control (Table 1).

Table 1. Crop options and imposed herbicide levels evaluated in fallowed sugarcane fields.

39

Preplant incorporated applications of trifluralin [2,6-dinitro-N, N-dipropyl-4-(trifluoromethyl)benzenamine] were made to designated fallow only and fallow + soybean plots on May 5, May 26, and April 20 and on wheat + fallow plots on June 16, June 16, and June 6 following wheat harvest and row buildup in studies A, B, and C, respectively. Trifluralin was selected for this study because it is labelled for use on soybeans and sugarcane and because it controls johnsongrass germinating from both seed and rhizome buds (6). Trifluralin was incorporated to a depth of 10 cm with a rotary tiller within 2 hours of application. To enhance the degree of broadleaf weed control with a minimum of soybean injury, metribuzin at 0.4 kg/ha was applied as a sequential preemergence treatment after trifluralin incorporation. All plots, regardless of herbicide level, were subjected to rotary tillage at the time of trifluralin application. In addition to doubling the rate of trifluralin, plots designated to receive the high level of herbicide usage also received two postemergence applications of sethoxydim[2-[l-(ethoxyimino)butyl]-5-[2-(ethylthio)propyl]-3-hydroxy-2-cyclohexen-l-one]at0.56 kg/ha on June 22 and July 6 (study A), July 18 and August 30 (study B), and June 17 and August 2 (study C) when johnsongrass was 60 to 152 cm tall. All herbicides were applied broadcast at 375 1/ha.

Where no herbicides were applied in the fallow only and wheat + fallow systems, weeds were controlled by plowing with an implement, consisting of a double lister plow and bed choppers, designed to open (15 cm depth) and reclose 1.8 m sugarcane rows in one operation. Fallow plowing was attempted at intervals of 3 to 4 weeks beginning after row build-up and rotary tillage of all plots within a management option. In the fallow only option, plowing was performed on June 16, July 6, and August 10, in study A. Early August rainfall limited this operation to June 16 and July 14 and June 6 and July 11 for studies B and C, respectively. Plots receiving no herbicide in the wheat + fallow option were plowed twice in study A (July 6 and August 10) and only once in studies B (July 14) and C (July 11). Since rainfall prevented a third tillage operation in mid-August in studies B and C, no additional tillage was attempted prior to the September ratings. Cultivation could not be performed in the fallow + soybean option. Therefore, metribuzin at 0.4 kg/ha was applied preemergence after planting, to represent the low level of herbicide usage, primarily to control broadleaf weeds.

Predominant weeds in these fields included: seedling johnsongrass, annual sedge, (Cyperus compressus L.), junglerice [Echinochloa colonum (L.) Link.], southern crabgrass [Digitaria ciliaris (Reta.) Koel.], red morningglory (Ipomoea coccinea L.), entireleaf morningglory (Ipomoea hederacea var. integruiscula Gray), cutleaf groundcherry (Physalis angulata (L.) St. Hil.], spotted spurge, (Euphorbia maculata L.), texas weed [Caperonia plaustris (L.) St. Hil.], and spreading dayflower (Commelina diffusa Burm. F.), with grasses being the predominant weed species.

Weed control efficacy for the various herbicide levels was determined in mid to late September by visually estimating the percentage of the surface area of each plot that could be covered by the foliage of each weed present (5). In effect, this was an estimate of the leaf area index of each weed species. Johnsongrass panicles on each plot row were also counted at this time.

The experiment was a three crop option by three herbicide level factorial arranged as a randomized complete block designed with four (studies A and B) or five (study C) replicates per treatment. Experimental plots were 5.3 m wide and 15.3 m long. Data for each study were subjected to statistical analysis and means were separated using Fisher's Least Significant Different Test.

RESULTS AND DISCUSSIONS

Crop yields. Wheat yields averaging 1196 and 3622 kg/ha were obtained in 1982 for Coker 68-15 and Florida 301, respectively. Differences in yield resulted from a severe infection of leaf rust (Puccinia recondita F. sp. tritici) and stem rust (Puccinia graminus F. sp. tritici) on Coker 68-15. In 1983, yields of 1740 and 1270 kg/ha were obtained for Coker 762 and Coker 797, respectively. During the pollination and early milk stages of wheat development in 1983 (March 15 to April 15), rain totalling 28.2 cm fell during 14 days of the period. These cloudy and wet conditions probably contributed to the low cultivar yields. In 1985, the short growing season caused by replanting resulted in a yield of 1492 kg/ha for Coker 747.

Soybeans were not harvested in these studies because the late maturity dates of current recommended soybean cultivars would have jeopardized sugarcane planting during the traditional planting period of late August and September. Delaying the planting of sugarcane would then cause conflicts with the sugarcane harvesting season. Soybeans would, thus, be better suited as a green manure crop turned under 4 to 6 weeks prior to

40

sugarcane planting. This would limit potential benefits from the use of soybeans unless an earlier maturing cultivar can be identified for this area.

Weed responses. Management options by herbicide level interactions were observed at the P<0.05 level in 1982 and 1983 and at the P<0.10 level in 1985 for total weed cover and johnsongrass panicle production. Interaction-indicating F-values of 24.75, 3.74, and 2.11 (total weed cover) and 27.04, 34.94, and 2.44 (johnsongrass panicle production) were obtained in 1982, 1983, and 1985, respectively. Similar interactions at the P<0.05 level were observed for johnsongrass foliar cover in 1982 and 1983, where F-values of 2.78 (1982) and 6.93 (1983) were obtained, but not in 1985, where an F-value of 1.33 was obtained. These interactions indicate that weed responses were generally a function of both the crop and herbicide management option selected for all parameters.

Where plowing represented the low level of herbicide usage in plots fallowed during the summer months (fallow only and wheat + fallow options), total foliar cover and johnsongrass foliar cover was lower in 1982 than in 1983 and 1985, apparently because plots received an additional plowing on August 10, 1982 (Table 2).

Table 2. Effects of crop-herbicide management options on total weed (TW) and johnsongrass (JG) foliar cover in fallowed sugarcane fields1.

1 Foliar cover is a visual estimate of the percentage of the surface area of each plot that could be covered by the foliage of each weed present, i.e. leaf area index by weed species.

Junglerice and the other small-seeded annual weeds germinated with johnsongrass in plots receiving no herbicide and only two plowings (studies B and C). Under such circumstances the vegetative development of johnsongrass was hindered; hence, its foliar cover was reduced by competition from these weeds. An interaction was observed because when soybeans were included with the low level of herbicide usage, small-seeded annual

41

weed development was suppressed more than johnsongrass development. As a result, total weed cover was lower and johnsongrass foliar cover in these plots was generally higher and often equalled that of plots that received only June and/or July plowings in the fallow only and wheat + fallow options.

From observations, the medium level of herbicide usage in the fallow only and wheat + fallow options had lower densities of weeds; however, escaped weeds grew larger, producing weed covers that equaled or exceeded those in plots receiving no August plowing (low level). When soybeans were included with the medium level of herbicide usage, this was not observed because the soybeans provided additional suppression of escaped weeds.

Total weed cover was greatly reduced at the high level of herbicide usage where the rate of trifluralin was increased and two postemergence applications of sethoxydim were included. Greatest reductions were observed in plots planted to soybeans in the spring and summer months. Although not completely eliminated, johnsongrass comprised only a small percentage of the total weed cover of these plots.

Plowed plots of the fallow only option generally had fewer johnsongrass panicles than plots receiving the medium level of herbicide usage with the difference being greatest where weather permitted three plowings (study A) (Table 3).

Table 3. Effects of crop-herbicide management options on johnsongrass panicle production in fallowed sugarcane fields1.

1 Johnsongrass panicles were counted on each plot-row in September of each year.

Where an August plowing could not be performed, junglerice and other annual weeds germinating simultaneously with johnsongrass probably restricted the subsequent reproductive development of seedling

42

johnsongrass as has been noted in other weed communities (4). As a result, johnsongrass panicle production in these plots generally equaled that of plots receiving a combination of soil-applied and postemergence herbicide treatments. Where soybeans were present during the summer months, johnsongrass panicle production generally decreased as the level of herbicide usage within the program increased, with little or no difference in johnsongrass panicle production between the medium and high levels of herbicide usage being observed. In this management option, weed-weed competition was replaced by crop-weed competition. Use of the preemergence herbicides (medium herbicide level) insured the development of a competitive soybean crop early in the growing season and resulted in increased johnsongrass suppression. As a result of this suppression, additional reductions in johnsongrass panicle production did not occur with the inclusion of two postemergence applications of sethoxydim to the management option.

In summary, an intensive chemical program employing a combination of broad-spectrum soil-applied herbicide(s) and multiple postemergence herbicide applications will be required to successfully insure a depletion of the soil reserves of johnsongrass seed and rhizomes over variable environmental conditions in fallowed sugarcane fields. If the program is not intensive and only soil-applied herbicides are used, control of small-seeded annual grass and broadleaf weeds would be expected to continue for a longer period than that of johnsongrass. As a result, emerging johnsongrass could grow and develop unencumbered for space by competition from other weed species and replenish the soil reserves of seed and rhizomes. Where a preemergence herbicide treatment for grass and broadleaf weeds (crop-herbicide management options employing fallow plowing only or multiple postemergence applications of herbicides) is excluded, germination of all weed species would begin at the same time and interspecific weed competition would result in reduced johnsongrass germination and development, but the depletion of johnsongrass seed reserves in the soil would be limited by the fact that the germination of dormant johnsongrass seed was not stimulated.

Soil cover provided by a wheat crop during the early months of the fallow period would probably be of benefit by inhibiting the germination of weed seeds and johnsongrass rhizomes located near the soil surface. Direct results of this inhibition would be a reduction in tillage requirements and a partial depletion of soil reserves of weed propagules resulting from enhanced natural mortality of the propagules while they remained in a quiescence state. However, once the wheat is harvested and the land disked, additional weeds will germinate and a control strategy for the summer months will still be needed.

With soybeans as a cover crop during the summer months, johnsongrass emergence could be suppressed to the point where the crop cover would eliminate the need for summer tillage and/or high (2X) rates of trifluralin and postemergence herbicide applications. To harvest the soybeans as a cash crop and integrate this practice into sugarcane culture, earlier maturing soybean cultivars must be developed and/or sugarcane planting delayed. By planting soybeans on reformed sugarcane beds instead of the conventional "flat-culture", this delay could be shortened because existing rows could be opened and sugarcane planted with a minimum of moisture-depleting seedbed preparation.

REFERENCES

1. Black, A. L. and J. F. Power. 1965. Effect of chemical and mechanical fallow methods on moisture storage, wheat yields, and soil erodibility. Soil Sci. Soc. Amer. Proc. 29:465-468.

2. Burnside, O. C, G. A. Wicks, and D. R. Carlson. 1980. Control of weeds in an oat (Avena sativa) -soybean {Glycine max) ecofarming rotation. Weed Sci. 28:46-50.

3. Fenster, C. R., and G. A. Wicks. 1982. Fallow systems for winter wheat in Western Nebraska. Agron. J. 74:9-13.

4. Radosevich, S. R. and J. S. Holt. 1984. Weed Ecology Implications For Vegetation Management. John Wiley & Sons. New York, 265 pp.

5. Richard, E. P., Jr. and L. M. Kitchen. 1989. Control of johnsongrass in fallowed sugarcane fields. Jour. ASSCT 8:12-18.

6. Standifer, L. C. and C. H. Thomas. 1965. Response of johnsongrass to soil-incorporated trifluralin. Weeds 13:302-306.

43

SUGARCANE RESPONSES TO Mn SOURCES AND S APPLICATION ON TWO FLORIDA HISTOSOLS

D.L. Anderson, Assoc. Professor, Everglades Res. & Ed. Ctr. University of Florida, P.O. Box 8003, Belle Glade

M.F. Ulloa, Agronomist, New Hope Sugar Cooperative, Pahokee, FL Fla. Agr. Exp. Sta. Journal Series No. 9399

ABSTRACT

In the Florida sugarcane industry, crop deficiencies of Mn are often observed and apparent yield declines are realized. To avoid these problems, furrow-applied Mn and elemental S are recommended at planting of sugarcane (Saccharum spp.) when soil pH exceeds 6.5. Sugarcane yield response to Mn and elemental S application were measured for three crops years on two Pahokee muck soils (Euic, hyperthermic Lithic Medisaprist). Two cultivars (cv. CP 72-1210 and CL 61-620) were grown, one at each location. The objectives were to examine the effectiveness of Mn and S recommendations determined from soil test information and to evaluate differences in the effectiveness of various Mn sources and elemental S on two specific soils with past histories of Mn deficiency and moderate to high soil pH. During six crop years, application of Mn had no significant effect on sugarcane yields, although application of elemental S for modification of soil pH had a moderate but inconsistent effect on yields at each location. Regardless of treatment and yield, the most significant finding was that soil pH increased after each cropping year, thus influencing the decline of leaf Mn concentration in each successive crop. These studies indicated that commercial plant application of Mn and elemental S may not always be justified under current soil testing recommendations on Florida Histosols.1

INTRODUCTION

Recommendations of Mn and S applications on sugarcane in Florida are based on soil pH determinations from the Everglades Soil Testing Laboratory (ESTL). When the soil pH is greater than 6.0, 5.6 kg Mn ha-1 (furrow-applied at planting) is recommended; when the soil pH is greater than 6.5, 575 kg S ha-1 is recommended for Histosols (4). A majority of the soils used for sugarcane production are Histosols. Most of these organic soils are located in the Everglades Agricultural Area (EAA), and are shallow soils overlying calcium carbonate bedrock and marl. Although soil pH's in the EAA range from 4.0 to 8.1, a vast majority of these soils have pH's greater than 6.0 and 6.5. Therefore, it is recommended for the majority of organic soils passing through the ESTL, that Mn and S be applied.

Manganese deficiency has been a well recognized barrier to attaining high-yield crop production in the EAA (2). The need for Mn has been recognized for crops such as sugarcane, vegetables, and rice grown on organic soils of high pH (1,6,7,16). High soil pH has been recorded as the most influential factor that reduces Mn availability for many crops in the EAA, including sugarcane (6,10,11).

When the soil pH is higher than 6.5, furrow application of elemental S is recommended at planting of sugarcane. This practice is recommended on the premise that localized pH changes in the furrow will increase micronutrient (i.e., Mn) plant availability, thus increasing yields. Sulfur content of EAA soils are sufficiently high to rule out S deficiency (12). This recommendation originated from only a very few agronomic tests in the past. Stevens (17) increased recoverable sugar by 10% by applying elemental S, but did not increase cane yield. Temporary alleviation of Mn deficiency during the plant crop was attained by Andreis and Gascho (6), in which cane and sugar yields were increased by 8% and 10%, respectively. In their study, soil application of elemental S was as effective in controlling Mn deficiency as applying various Mn sources, but only during the first year of

1Mention of a trade name, commercial product, or commercial enterprise does not constitute endorsement by the University of Florida.

44

production. The objectives of the studies reported herein are: to test the validity of Mn and S recommendations on high pH organic soils used for sugarcane production; and to investigate the responses of sugarcane to various Mn sources and to S application.

METHODS AND MATERIALS

Two locations approximately two km apart in the east central EAA were used. The soil at both locations was a Pahokee muck soil (Euic, hyperthermic Lithic Medisaprist), a Histosol that accounts for about 27% of the EAA under cultivation (personal communication, S. McCollum, Soil Conservation Service, Palm Beach Co., FL). Sources and treatments of Mn and S used at each location are listed in Table 1. The treatment design was a randomized complete block, using four replications.

Table 1. Treatments and sources of Mn and S used at each location.1

1 Cultivars CP 72-1210 and CL 61-620 used one at each location.

2 Included only at location two using CL 61-620.

3 IPM minus Mn used for all Mn and S source treatments.

The experimental areas were prepared for sugarcane planting by rototilling and furrowing so that each plot contained four rows on 1.5 m spacings by 10 m long. At each location, soil samples were taken and submitted for analyses at the Everglades Soil Testing Laboratory, EREC, Belle Glade. Soil test values for pH, Pw, and K respectively were 6.73, 2.6 kg ha-1, and 60 kg ha-1 at location 1, and 7.46, 1.8 kg ha-1, and 30 kg ha-1

at location 2. Based upon soil test information, P, K, Cu, Fe, Zn, and B were applied at planting in the furrows of all plots at 37, 100, 2.2, 2.2, 2.2, and 1.1 kg ha-1, respectively. Double lines of sugarcane stalks were cut to approximately 46-cm lengths and placed in the furrows and covered. The times of planting, soil and tissue sampling, and harvesting for both locations are given in Table 2. Locations one and two were planted using sugarcane cv. CP 72-1210 and cv. CL 61-620, respectively. All cultural practices were the same as those maintained by the grower. Soils were sampled (0-15 cm depth) within each treatment plot, between planted rows of cane. Fifteen top visible dewlap (TVD) leaf blades, with mid-ribs (18), were collected from each plot from both locations during June of each growing season (Table 2). Analysis for Mn concentrations in the TVD leaf blades was determined by the dry ash combustion method (5).

45

Table 2. Times of planting, sampling of soil and leaf tissues, and harvesting for both cultivar locations.

Planting

Harvest

Soil

TVD Leaf

Crop Age (mo.)

Plant 1st Ratoon 2nd Ratoon

Plant 1st Ratoon 2nd Ratoon

Plant 1st Ratoon 2nd Ratoon

Plant 1st Ratoon 2nd Ratoon

30 October 1983

19 February 1985 5 February 1986 16 December 1986

21 October 1983 15 March 1985 14 March 1986

16 June 1984 5 June 1985 25 June 1986

17.0 12.5 11.2

26 October 1984

6 March 1986 6 January 1987 21 November 1987

6 December 1984 18 March 1986 7 May 1987

22 July 19851

23 June 1986 2 July 1987

17.7 10.9 11.4

1 Sampled and analyized by Dr. R. Illey, Applied Agricultural Research, Lakeland, FL.

From the soil test pH at both locations, both 2.2 kg Mn ha-1 and 560 kg elemental S ha-1 were recommended for furrow application at planting (4). Cultivar CP 72-1210 was used because it is used by the majority of sugarcane industry in Florida; cv. CL 61-620 was also used because it typically exhibits chlorotic symptoms indicating Mn deficiency early in the growing season, especially at high soil pH.

At the time of harvesting (Table 2), cane was burned to remove excess leaves and trash, and whole stalks were cut by hand at the soil surface. Tops were removed by cutting at the top hard internode. After the sugarcane stalks from each plot were weighed, 15 stalks per plot were randomly collected and passed through a three-roller sample mill for juice extraction. The crusher juice was analyzed for Brix using a refractometer (Bausch & Lomb, Inc., Rochester, NY). The juice was clarified using lead subacetate (13, p. 54), and the Pol was determined using a Rudolph Autopol IIS automatic saccharimeter (Rudolph Research, Flanders, NJ). The percent sucrose in juice was estimated using formulas developed from sucrose Pol and Brix temperature correction (CBrix) tables given by Meade and Chen (13, p. 882-885; and p. 861-862, respectively):

Sucrose = (Pol x 26)/{105.811 + [(CBrix - 150) x 0.44]}; 1)

where the 20°C temperature correction for Brix is given as:

CBrix = Brix + [(temperature - 20) x 0.075]. 2)

juice purity was calculated as the percent of the ratio of sucrose to CBrix. Recoverable 96° sugar (STC, kg sugar Mg'1 cane) was calculated using the formula described by Rice and Hebert (14):

STC = [(Sucrose x 21.058) - (CBrix x 6.15) x VCF]; 3)

where VCF is the varietal correction factor of 0.965 for cv. CP 72-1210 and assumed 1.00 for CL 61-620. From the measured tons cane per hectare (TCH, Mg cane ha-1) and STC, the tons sugar per hectare was calculated (TSH, Mg 96° sugar ha-1). Analyses of variance (ANOVA) and standard error of regression (SER) were obtained using SAS (15).

46

Event Crop CP 72-1210 CL 61-620

RESULTS AND DISCUSSION

Significant (P>0.05) declines in stalk weight, TCH, and TSH were observed at each location and after each successive crop (Table 3). No cropping year trends were observed for CBrix, sucrose, nor STC. Despite high soil pH values, there were no crop responses to any of the Mn sources at either location, although statistically significant (P>0.05) yield responses were measured as a result of S application (Table 4). Responses to elemental S application were not high for either cultivar. Cane yield (TCH) increased 3.6 and 6.8% during the same season, 1st ratoon at location one (cv. CP 72-1210) and during the plant crop at location two (cv. CL 61-620), respectively (Tables 2 and 4). Similar yield increases have been reported by others (6,17). Assuming the current price of elemental S, the application of 560 kg of elemental S per ha, and a return of $7-10 per metric tonne of cane, the observed yield increases would not economically justify applications of elemental S (3).

Table 3. Average yield data from both cultivar locations.

Location one Location two Yield Factors1 1 2 32 LSD.05 1 2 3 LSD.05

3

CBrix (%) Sucrose (%) Stalk (kg) TCH STC TSH

21.60 20.40 1.57

136.00 143.00 19.50

21.40 19.60 1.47

123.00 136.00 16.80

20.10 19.20 1.26

105.00 135.00 14.10

0.20 0.20 0.03 3.00 2.00 0.50

18.20 17.60 1.39

121.00 130.00 15.70

19.80 18.90 1.33

114.00 139.00 15.90

20.30 18.20 1.10

82.00 129.00 10.70

0.20 0.20 0.06 2.00 2.00 0.40

1 TCH = cane yield (Mg cane ha-1), STC = sugar per ton of cane (kg Mg-1), and TSH = sugar yield (Mg ha-1). 2 Plant crop=l, 1st ratoon=2, and 2nd ratoon=3.

3 LSD.05 between crops and by location.

Table 4. Significant yield responses to S application.

Location one Location two S - 2 1 - 1 -2-

Applied TCH2 Stalk TCH TSH Stalk

kg/ha Mg/ha Kg Mg/ha Kg 0 118.5 1.34 118.5 15.40 1.31

280 125.2 1.35 120.1 15.48 1.35 560 122.8 1.47 126.6 16.13 1.43

LSD.05 6.5 0.12 3.1 0.58 0.13

Mean 123.4 1.38 121.2 15.68 1.33

1 plant crop = l, 1st ratoon=2, and 2nd ratoon=3.

2 TCH=cane yield, TSH=sugar yield, and Stalk-stalk weight.

47

At both locations, soil pH increased significantly during each crop year of sugarcane cultivation (Table 5). Although the data presented herein do not explain these observations, this occurrence is common among organic soils in the EAA in which there is high evapotranspiration bringing Ca salts to the soil surface, and in which high pH, carbonate-saturated irrigation water is used (9,12). Across all crop years and for both locations, the relationship between soil pH and TVD Mn content was significant (P>0.01); as soil pH increased, Mn in TVD leaf tissues declined (Table 6). This observation was also made by others (6,10). At both locations, by the third year of production, the TVD Mn concentration approached the critical nutrient level of 10 ppm (8) as pH > 7.5 (Figure 1). Unlike studies by others (6,10), none of the Mn or S treatments had any effect on tissue content of Mn in these studies. In studies conducted by Andreis and Gascho (6), the soil pH was raised by application of limestone in order to induce Mn deficiency. However, due to the high buffer pH capacity of the soil they used, the soil pH declined with time - the reverse of our observations. Although not discussed by Andreis and Gascho (6), this may explain why they did not obtain crop response to Mn and S after the plant crop.

Table 5. Soil pH changes observed at both locations during three crop years of cultivation.

- - Soil pH

Location one Location two Source2 l1 2 3 1 2 3

1 Plant crop=l, 1st ratoon=2, and 2nd ratoon=3. 2 IPM and Control treatments not included at this location. 3 LSD and mean across all crops.

48

Control IPM

MnO G

STM5 SG

0 S 280 S 560 S

L S D . 0 5

Mean

6.80 6.75 6.70 6.70

6.80 6.67 6.67

0.13

6.73

7.12 7.18 7.22 7.18

7.30 7.05 7.18

0.13

7.17

7.85 7.98 7.72 7.78

8.02 7.98 7.70

0.28

7.86

0.083

7.26

7.52 7.48

7.40 7.48 7.38 7.48

7.62 7.48 7.35

0.09

7.46

8.22 8.15

8.12 8.18 8.18 8.18

8.18 8.20 8.18

0.10

8.17

8.27 8.20

8.27 8.25 8.25 8.27

8.27 8.27 8.18

0.09 0.04

8.25 7.95

Table 6. Leaf TVD Mn concentrations at both locations during three crop years of cultivation.

Control IPM

MnO G

STM5 SG

OS 280 S 560 S

LSD. 0 5

Mean

..

82 68 70 59

73 68 68

18 70

32 29 32 32

32 29 32

9 31

15 14 15 15

14 16 14

3 14

34 35

35 31 33

35

34

20 22

18 22 20 20

17 23 25

7 21

13 13

13 104

12 12

14 15 14

5 13

Plant crop = l, 1st ratoon=2, and 2nd ratoon = 3. 2 IPM and Control treatments not included at this location. 3 Sampled without midrib; analysis by Dr. J.R. Illey, Applied Agricultural Research, Lakeland, FL 33801;

some data missing, thus mean and LSD not calculated. 4 10 ppm is critical nutrient level (8).

Figure 1. Effect of soil pH on sugarcane TVD leaf MN concentration from two locations and three crop years.

49

.....Leaf TVD Mn (ppm) Location one Location two

Source2 l1 2 3 l3 2 3

CONCLUSIONS

At the locations studied, soil pH increased after each cropping year, thus influencing the decline of leaf Mn concentrations of successive crops. Although recommendations for Mn and elemental S application were given according to initial soil pH's greater than 6.5, applications had little to no effect on sugarcane yields and no effect on Mn concentrations found in TVD leaf tissues. It was therefore recognized that soil pH may not be a sufficient criteria for recommending Mn and S. Perhaps the most significant finding was that after each crop year of sugarcane cultivation soil pH increased, which resulted in significant decline of Mn content in TVD leaf tissues. These results indicate that there may be greater potential and justification to correct for Mn deficiencies in the ratoon crop, instead of at planting.

ACKNOWLEDGEMENTS

The authors would like to express their appreciation to management and field personnel of New Hope Sugar Cooperative, whom without their support, these studies would not have been possible. Appreciation is also expressed to Wedgeworth's, Inc., Belle Glade, FL for their contribution of fertilizer materials, and to E. Bussey, C. Miller, and N. Relph for their technical assistance.

REFERENCES

1. Allison, R.V. 1932. The use of the less common elements as soil amendments for sugarcane production in South Florida. Proc. 4th Cong. Int. Soc. Sugar Cane Technol., San Juan, Puerto Rico. Bull. No. 112.

2. Allison, R.V., O.C. Byran, and J.H. Hunter. 1927. The stimulation of plant response on the raw peat soils of the Florida Everglades through the use of copper sulfate and other chemicals. Fla. Agr. Exp. Sta. Bul. 190.

3. Alvarez, J., and F. Rohrmann. 1985. Costs and returns for sugarcane production on muck soils in Florida, 1983-84. Econ. Information Rep. 204. Food Res. Econ. Dep., Coop. Ext. Ser., Univ. FL, Gainesville 32611. 18 pp.

4. Anderson, D.L. 1985. Crop soil fertility recommendations of the Everglades Soil Testing Laboratory. Mimeo Report EV-1985-10. EREC, Univ. FL, Belle Glade, FL.

5. Anderson, D.L. and L.J. Henderson. 1988. Comparing sealed chamber digestion with other digestion methods used for plant tissue analysis. Agron. J. 80:549-552.

6. Andreis, HJ. and GJ. Gascho. 1971. Improving manganese nutrition of sugarcane grown on high pH organic soils. Soil Crop Sci. Soc. FL Proc. 31:110-113.

7. Beverly, R.B. 1984. Managing high soil pH and Mn availability in Everglades vegetable production. EREC Res. Rep. 11. Nov. 17 pp.

8. Bowen, J. E. 1983. Micro-element nutrition of sugarcane. 3. Critical micro-element levels in immature leaf sheaths. Trop. Agric. (Trinidad) 60(2): 133-138.

9. Burdine, H.W. and V.L. Guzman. 1968. Some soil pH effects on soil test and growth on some vegetable crops on Everglades organic soils. EREC Res. Rep., EV 68-9. Univ. FL, Belle Glade.

10. Gascho, GJ., H.W. Burdine, and FA. Taha. 1971. Availability of manganese in Everglades peat as influenced by soil pH, source, and rate of application. Soil Crop Sci. Soc. FL Proc. 31:106-109.

11. Gascho, GJ., M. Isobe, and HJH. Hagihara. 1977. Manganese availability in sugarcane soils of Hawaii and Florida. Proc. ISSCT 16:971-982.

50

12. Lucas, R.E. 1982. Organic Soils (Histosols). Formation, distribution, physical and chemical properties and management for crop production. Res. Rpt. 435 Farm Sci. Michigan State Univ. Agric. Exp. Sta. and Coop. Ext. Ser., East Lansing.

13. Meade, G.P. and J.C.P. Chen. 1977. Cane Sugar Handbook. 10th ed. John Wiley & Sons, New York.

14. Rice, E.R. and L.P. Hebert. 1972. Sugarcane variety tests in Florida during the 1971-72 season. USDA Agricultural Research Service S-2. U.S. Government Printing Office, Washington, D.C.

15. SAS Institute Inc. 1985. SAS procedures guide for personal computers. Version 6 ed. Cary, NC. 373 pp.

16. Snyder, G.H., D.B. Jones, and C.L. Elliot. 1986. Amelioration of pH-induced Mn deficiency in Histosol grown seedling rice. Agron. Abstr. p. 216. ASA, New Orleans, La.

17. Stevens, F.D. 1933. Agronomic phases of sugarcane investigations. Fla. Agr. Exp. Sta. Ann. Rep. p 184-185.

18. Thein, S. and G J. Gascho. 1980. Comparison of six tissues for diagnosis of sugarcane mineral nutrient status. Proc. ISSCT 16:152-163.

51

EFFECT OF WHITE GRUB (LIGYRUS SUBTROPICUS (BLATCHLEY)) INFESTATIONS ON SUGARCANE ROOT:SHOOT RELATIONSHIPS1

F. J. Coale and R. H. Cherry University of Florida

Institute of Food and Agricultural Sciences Everglades Research and Education Center

Belle Glade, Florida

ABSTRACT

The white grub Ligyryus subtropicus (Blatchley) is a major insect pest of Florida sugarcane (Saccharum spp.). The primary impact this insect has on the sugarcane plant is through larval feeding on plant roots and underground stems. A greenhouse study was conducted to evaluate the effect of L. subtropicus feeding on the relative growth and development of sugarcane stalks and root systems. Third instar L. subtropicus larvae were introduced to the soil at the rate of 0,2,4,6, and 8 larvae/plant. Overall grub survival in all infestation levels at the end of the experiment was approximately 86%. Root and stool (underground stem) dry matter were reduced linearly by increased grub infestation. Above ground dry matter was not affected by grub feeding. Root:shoot dry matter ratios decreased as number of grubs/plant increased. Juice extraction and juice quality at time of harvest was not affected by the number of grubs/plant.

INTRODUCTION

Sugarcane (a complex hybrid of Saccharum spp.) is Florida's most valuable field crop, and is primarily grown in the Everglades Agricultural Area of southern Florida. Ingram et al. (5) first reported white grub injury to Florida sugarcane in 1938. Among the several white grub species currently associated with sugarcane in Florida, Ligyrus subtropicus (Blatchley) is the species of primary economic importance (4). Currently, no chemical control is known for this pest and flooding of sugarcane fields is sometimes necessary for control (2). The primary impact this insect has on the sugarcane plant is through larval feeding on the roots and underground stems. Miller and Bell (6) reported a study in which L. subtropicus adults were placed in buckets containing young sugarcane plants. As the resulting larvae developed, for each gram increase in weight of the larvae, there was a corresponding decrease in plant root weight of 12.8 g. Field observations have shown that L. subtropicus infested sugarcane is stunted, chlorotic, and easily lodged. Sugarcane and sugar yields may be severely reduced as the result of white grub injury. Sosa (8) described a sugarcane yield reduction of 28% and a sugar yield loss of 39% attributable to a L. subtropicus infestation difference of 12 versus 1 grub/m row. In the same study (8), a 73% reduction in the ability of the crop to ratoon under high white grub infestation was reported.

Little information is available on the effect of white grub feeding damage to sugarcane prior to crop harvest. Yield reductions often result from stalk lodging and recumbent growth (3), which are secondary effects of white grub feeding damage. Grub destruction of plant root systems and underground supportive and regenerative tissue prior to stalk lodging may affect the potential productivity of the sugarcane crop. The objective of this study was to evaluate the effect of L. subtropicus feeding on root system destruction and sugarcane plant development prior to stalk lodging.

MATERIALS AND METHODS

Sugarcane (var. CP 72-1210) was planted in a field on Pahokee muck (Euic, hyperthermic Lithic Medisaprists) on 9 February 1987, at Belle Glade, Florida. On 6 August 1987, intact sugarcane plants were transplanted into 19 L plastic buckets by removing cylindrical soil cores (25 cm diameter, 35 cm deep) containing an undisturbed sugarcane plant. The buckets were transported to a greenhouse and the plants were maintained for 39 days. Third instar L. subtropicus grubs, collected from commercial sugarcane fields, were then introduced

'Florida Agricultural Experiment Station Journal Series No. 9568.

52

to the buckets at levels of 0, 2, 4, 6, and 8 grubs/bucket. The experiment was a randomized complete block design with 8 replications. After 42 days, the stalks were cut at the soil surface and the number of stalks/bucket and total fresh weight per bucket were recorded. Stripped stalk fresh weight and stalk height to the top visible dewlap leaf position were determined. Stalks were crushed in a 3-roller extraction mill and the crusher juice and bagasse were collected and weighed. Crusher juice brix was determined with a laboratory refractometer. The extracted juice was filtered with lead subacetate to remove suspended solids. Sucrose concentration in the filtered juice was determined with a polarimeter. Sugar concentration, expressed as g sugar/kg cane, was calculated using Arceneaux's modification of the Winter-Carp-Geerligs formula (1). Soil was gently washed from the root system and stool with a water spray. Surviving grubs were collected and root segments that had been severed by grub feeding and were no longer physically connected to the plant were discarded. Attached roots were cut from the stool, thoroughly washed with water, and retained. All plant tissue was dried to constant weight at 60° C.

RESULTS AND DISCUSSION

The number of grubs/bucket did not affect grub survival rate and the overall survival at all infestation levels during the experiment was approximately 86%.

Table 1 presents the effect of white grub infestation on root dry matter. These root mass data represent only the remaining intact and potentially functional portion of the plant's total root system. Observations indicated that actual root mass consumption was a minor contribution to the total damage inflicted by the grubs. The major impact grub feeding had on the root system was through severing roots from the stool. Root function may have been severely impaired due to the large number of severed roots. Root dry matter/plant decreased linearly with increasing number of grubs/plant. The 8 grubs/plant treatment resulted in a 59% reduction in functional root mass versus the control.

Table 1. Effect of white grub infestation on root, stool, and aerial shoot dry matter (DM) of sugarcane.

Grubs/Plant Root DM Stool DM Shoot DM

mg/plant g/plant g/plant

** Regression is significant at P<0.01. Root DM = 540 - 29(Grubs/Plant), r2 =.44. Stool DM = 1.94 - 0.06(Grubs/Plant), r2 =.58.

NS Regression is non-significant (P>0.05).

A sugarcane stool is composed of supportive and regenerative underground stems and roots. Stool dry matter data reported in this study refers only to underground stem tissue. Stool dry matter decreased linearly with increasing number of grubs/plant (Table 1). The 8 grubs/plant treatment resulted in a 31% decrease in stool mass versus the control. This reduction in stool mass was highly localized and resulted in completely severed or highly weakened stalk bases leading to stalk lodging. Feeding damage to stool tissue may have a very

53

0 2 4 6 8

435 466 379 399 178

1.60 1.48 1.24 1.34 1.11

322 347 297 322 310

Linear Regression ** ** NS

pronounced effect on the regenerative capacity of the plant due to the destruction of underground axillary buds. Sosa (8) reported a 73% reduction in the stubbling ability of sugarcane highly infested with L. subtropicus grubs.

During this experiment, total aerial shoot dry matter accumulation was not significantly affected by changing grub densities (Table 1). After 42 days of exposure to the highest grub infestation rate (8 grubs/plant), sugarcane plants were exhibiting leaf chlorosis and stalk lodging. At the onset of these visual plant stress symptoms, severe destruction of subterranean plant tissue had already taken place, but aerial shoot dry matter production was not reduced. It is probable that if the experiment had been continued after the 42 days, aerial production would have been severely impaired due to the reduced capacity of the plant's root system and supportive structures to sustain shoot growth. Data in Table 1 show that the severity of grub damage to the sugarcane plants was roots > stool > shoots.

Table 2. Effect of white grub infestation on root:shoot dry matter relationships of sugarcane.

Grubs/Plant Root:Aerial shoot Subterranean:Aerial shoot1

mg/g mg/g

1 Subterranean includes root and stool. ** Regression is significant at P<0.01.

Root:Aerial shoot = 1.48 - .08(Grubs/Plant), r2 = .48. Subterranean:Aerial shoot = 6.95 - .23(Grubs/Plant), r2 = .48.

The effect of white grub infestation on root:shoot dry matter relationships is presented in Table 2. The relative quantity of functional root mass supporting the growth of the above ground plant decreased linearly with increasing number of grubs/plant. The ratio of total subterranean dry matter to total aerial dry matter also decreased linearly with increasing grub rate.

At the termination of this experiment, the sugarcane plants had not attained harvest maturity. This fact notwithstanding, the plants were handled as if mature millable stalks were present. Regression analyses indicated that grub infestation level did not significantly affect millable stalk fresh weight, stalk length, or crusher juice extraction. Both percent Brix and percent sucrose were low (13.5 and 10.7, respectively) and not significantly different across the range of grubs/plant treatments. The low crusher juice Brix and sucrose analyses reflected the immaturity of the milled stalks (7). Sugar concentration was also unaffected by the level of grub infestation and averaged 67.5 g sugar/kg cane. These data are in contrast with Sosa's (8) analysis of mature, field-grown cane in which a three percentage point decrease in crusher juice extraction, a two percentage point reduction in juice sucrose, and a 17% reduction in sugar concentration were disclosed due to high L. subtropicus grub infestation levels. In our study, the lack of harvested crop response to grub feeding pressure was most probably due to the short duration (42 days) that the plants were exposed to the grubs.

The objective of this experiment was to evaluate the effect of L. subtropicus feeding on root system destruction and sugarcane plant development prior to stalk lodging. At the onset of the visual plant stress symptoms of leaf chlorosis and stalk lodging due to grub feeding activity, underground plant components had already been severely damaged. Above ground shoot development and sugar production had not been affected by L. subtropicus infestation at this time. It is suspected that, due to the drastic root system and stool destruction

54

0 2 4 6 8

1.29 1.29 1.22 1.26 0.55

6.32 5.69 5.34 5.63 4.02

Linear Regression ** **

observed in this study, significant reductions in above ground plant growth, cane production, and sugar yield would have been observed under high grub infestation levels if the experiment had been continued after the onset of stalk lodging. Therefore, it appears that the detrimental effects of grub feeding damage on juice quality and sugar yield, as reported by Sosa (8), are expressed during the later phases of crop development and maturation, i.e. after the crop has lodged.

REFERENCES

1. Arceneaux, G. 1935. A simplified method of making theoretical sugar yield calculations. In accordance with Winter-Carp-Geerligs formula. Int. Sugar J. 37:264-265.

2. Cherry, R.H. 1984. Flooding to control grub Ligyrus subtropicus (Coleoptera: Scarabaeidae) in Florida sugarcane. J. Econ. Entomol. 77:254-257.

3. Gascho, G J. and S.F. Shih. 1981. Cultural methods to increase sucrose and energy yields of sugarcane. Agron. J. 73:999-1003.

4. Gordon, R.N. and D.M. Anderson. 1981. The species of Scarabaeidae (Coleoptera) associated with sugarcane in south Florida. Fla. Entomol. 64:119-138.

5. Ingram, W., H. Hayes, and R. Lobdell. 1938. Sugarcane pests in Florida. Proc. ISSCT 6:89-98.

6. Miller, J.D. and M.G. Bell. 1983. Life cycle of the white grub and its effect on sugar cane. Jour. ASSCT 2:85.

7. Miller, J.D. and N.I. James. 1978. Maturity testing of sugarcane. Proc. ASSCT 7(NS):101-106.

8. Sosa, O., Jr. 1984. Effect of white grub (Coleoptera: Scarabaeidae) infestation on sugarcane yields. J. Econ. Entomol. 77:183-185.

55

THE RESPONSE OF SUGARCANE SELECTIONS TO SUGARCANE BORER IN THE GREENHOUSE AND FIELD

W.H. White and J.W. Dunckelman Sugarcane Research Unit, Agricultural Research Service

U.S. Department of Agriculture Houma, Louisiana, 70361

ABSTRACT

A greenhouse procedure, using plants regenerated from bud-chips, was evaluated as a means of screening sugarcane (Saccharum interspecific hybrids) selections for resistance to sugarcane borer (Diatraea saccharalis). In one experiment, ten cultivars were infested in the greenhouse with first-instar larvae of the sugarcane borer, damage expressed as percent deadhearts was determined 4 weeks later. These percentages in themselves did not show differences in resistance for 9 of the 10 cultivars, but when the percentages were rated from low to high, the position of the cultivars in rank generally agreed with the relative resistance categories determined under natural infestation at the outfield stage of the USDA sugarcane selection program in Louisiana. In a second experiment, 123 sugarcane selections were first screened using greenhouse procedures and then transplanted to the field and exposed to natural sugarcane borer infestation in both plant cane and first stubble. Correlation of percentages of deadhearts in the greenhouse and bored internodes in the field were for the greenhouse test versus plant cane crop [r=.16 (P=0.07)], first stubble crop [r=.14 (P=0.12)], and the mean of plant and first stubble crops [r=.17 (P=0.05)]. Correlation of percent internodes bored between plant cane and first stubble was moderate [r=.43 (P < 0.01)]. A cluster analysis using data from the three data sets revealed a group of 23 selections that were among the most susceptible and a smaller group of 12 selections that were among the most resistant in the three data sets. The results of this study suggest a relationship between the percentages of deadhearts from the greenhouse and internodes bored from the field; however, the results indicate that the reaction of cultivars measured by deadhearts in the greenhouse must be confirmed by field data before it can be used safely in selection.

INTRODUCTION

In Louisiana, the development of a new sugarcane cultivar usually requires 13 years and involves the evaluation of initial populations as large as 100,000 seedling (2). Selections are based on some 26 evaluation criteria, many of which require detailed observations and involved procedures. Because of the time, land area, and resources required, some evaluations must be delayed until the population of candidate cultivars has been drastically reduced by selection. One such evaluation, which requires detailed studies for quantitative information, is the reaction of selections to the sugarcane [Diatraea saccharals (F.)]; the most important insect pest in Louisiana sugarcane.

For more than 25 years, candidate cultivars in Louisiana have been evaluated for their response to sugarcane borers at the outfield stage of selection where only a few candidate cultivars are evaluated in larger, replicated plots. Although sugarcane borer data collected in the outfield have been useful in the management of released cultivars, such data rarely influenced the decision to release a cultivar.

Data on sugarcane borer resistance have been obtained a few years earlier in the selection program, infield stage (9), but efforts to obtain data earlier than this have met with limited success. Pan and Hensley (6) evaluated seedling progeny from several sugarcane crosses for resistance to sugarcane borer by infesting plants with first-stage larvae. Their technique appeared useful as a means of screening large numbers of seedlings, but the performance of the selected seedlings were not evaluated in field trials. Dunckelman and Legendre (4) reported that clones selected for sugarcane borer resistance as seedlings in the greenhouse were no more resistant to sugarcane borer damage than those not selected.

The purpose of this study was to evaluate greenhouse procedures for determining the response of potential new cultivars to sugarcane borer, some 3 years after initial selection, when the number of candidate cultivars has been reduced to a manageable level.

56

METHODS AND MATERIALS

During the 1984 harvest season, two stalks from each of 123 selections, five stalks from the cultivars CP 48-103, CP 61-37, CP 67-412, CP 70-330, CP 72-356, CP 72-370, and NCo 310, and 75 stalks from the cultivars CP 65-357, CP 70-321 and CP 74-383 were stored (10° C) for about 4 weeks before bud-chips were excised using procedures of Ramaiah et al. (7). All bud-chips were immersed in the fungicide benomyl-[methyll-(butylcarbamoyl)-2-benzimidazolecarbamate] (0.33 g ai/1) for 10 min. before being planted in vermiculite in 50 X 35 X 8 cm metal flats. Two experiments were established at this time; one consisting of only cultivars and the other with the 123 selections and three cultivar standards.

Evaluation of Cultivars. This experiment contained the following 10 cultivars: NCo 310, CP 48-103, CP 61-37, CP 65-357, CP 67-412, CP 70-321, CP 70-330, CP 72-370, CP 72-356, and CP 74-383. The experiment was replicated five times and the cultivars randomly assigned to one of the two flats comprising a replication. Six bud-chips per cultivar, planted in a row, were considered a replication.

Flats were placed in a greenhouse, with a daily temperature fluctuating between 10-32° C and fertilized at bi-weekly intervals with an all-purpose water soluble plant food; concentrated to 15-30-15. When plants were about 30 cm tall, each plant was infected with four ( + /- 2) laboratory-reared first-instar, sugarcane borer larvae by means of a hand-held inoculator (3, 10). Each plant was re-infested the following week with four larvae. Counts were made 4 weeks later to determine the total number of tillers and the number of sugarcane borer-killed tillers (deadhearted) for each cultivar. Data were expressed as percent deadhearts and percentages were transformed to arcsine V% for analysis as a randomized complete block design.

Evaluation of Selections. In greenhouse evaluations, bud-chips were used in a manner similar to the cultivar test. Six bud-chips from each of 123 selections were planted in flats with each flat containing four selections, and three cultivars serving as standards; the test was replicated twice. The cultivar standards and their relative resistance ratings (based on outfield cultivar trials) were as follows: CP 74-383, highly susceptible; CP 65-357, intermediate in resistance; and CP 70-321, highly resistant. Entries were randomly assigned to each flat and to location with a flat. Plants were handled identically to plants in the cultivar test with respect to both infestation and evaluation procedures.

Flats used in the greenhouse evaluation were sprayed with the insecticide monocrotophos (dimethyl-(E)-l-methy-2-methylcarbonylvinly phosphate] (0.84 kg ai/ha) to kill sugarcane borers, and individual plants were transferred to 7.5 cm peat-pots. Plants were then transplanted to the field in the spring to single-row plots 1.8 m long on rows 1.8 m apart in a randomized complete block design with two replications. Plants within a row were spaced 30 cm apart for a total of six plants per plot.

During 1985, plants were exposed to naturally occurring sugarcane borer. At harvest, all selections were evaluated in the plant cane crop for damage by randomly selecting 10 stalks from each plot and determining the number of internodes bored. The first ratoon crop was evaluated under natural sugarcane borer infestation. Data were taken in the same manner as the previous year and data from both crops were expressed as percent internodes bored.

Observed percent deadheart data from the cultivar standards in the greenhouse evaluation were tested for goodness to fit to the binomial distribution by chi-square. Probabilities for the binomial distribution were obtained from Beyer (1).

Cluster analysis (5), a multivariate statistical method, was used to combine the selection data from all three evaluations into a common data base for assigning selections into distinct groups. The FASTCLUS procedure (8) was used to perform a four cluster grouping on the mean damage response from the greenhouse, plant cane and first stubble field evaluations. A four cluster groupmg was chosen apriori to create intermediate categories of resistance.

RESULTS AND DISCUSSION

The results of the commercial cultivar test are shown in Table 1. The cultivars, except for CP 70-330, were not significantly different from each other, indicating very high variability in the greenhouse results. When

57

the cultivars are arranged in order of their percentages from greatest percentage of deadhearts to least, those with the highest, intermediate and lowest percentages are also those that rated similarly in outfield cultivar trials, with the exception that CP 48-103 which rated resistant in the greenhouse trial, rated susceptible in outfield cultivar trials. These data suggest a relationship in reaction of cultivars expressed as deadhearts in the greenhouse and internodes bored in the field.

Table 1. Percent deadhearts for commercial sugarcane cultivars evaluated in the greenhouse and their assigned resistant category based on outfield evaluations.

Resistance Variety Deadhearts1 category2

CP 61-37 CP 72-356 CP 74-383 CP 72-370 CP 65-357 NCo 310 CP 48-103 CP 70-321 CP 67-412 CP 70-330

(%)

93 a 90a 89 a 89 a 78 a 75 ab 67 ab 66 ab 57 ab 43 b

S S S IR IR R S R R R

1 Numbers in the same column not followed by the same letter differ significantly (P<0.05) as determined by the Student-Neuman-Kuels Range on the arcasine V% transformation.

2 Based on data from outfield cultivar trials.

Further evidence for the variability of greenhouse results was obtained from the three cultivars used as standards in the selection test. The three cultivars were planted in all 62 flats. It was decided to use only those data for each cultivar where all six bud chips had germinated, eliminating variable germination among replication which may have affected the results in Table 1.

The mean percent deadhearts and 95% confidence intervals for the 44 flats of CP 74-383 with full germination was 87.7 +/ 4.9, for the 38 flats of CP 65-357 was 67.0 +/-9.3 and for the 59 flats of CP 70-321 was 68.7 +/7.2.

Frequency distributions for the proportion of deadhearts for each cultivar standard were determined in an effort to identify the sources of variability. Frequency distributions were generated by determining the number of occurences for each discrete category. The discrete categories were zero among the six plants with deadhearts up to six of six plants with deadhearts.

The frequency distribution for the susceptible standard, CP 74-383, statistically fit a binomial (X2 = 0.51, P > .05, df = 2) while the distributions of the two resistant cultivars were bi-modal with highly significant Chi-square values when the observed frequencies were compared to those expected from the binomial distribution. The binomial distribution for CP 74-383 suggests that the probability of a plant of CP 74-383 becoming deadhearted was a random event. The bi-modality of the more resistant standards suggest that their becoming deadhearted was not random and that factors other than simply resistance were operating in the greenhouse. These factors, possibly experimental design, resulted in some cultivar reacting as resistant and others as susceptible.

58

Correlations for the selected cultivars between the percentages of deadhearts in the greenhouse and internodes bored in the field were r = .16 (P = 0.07), r = .14 (P =0.12), and r = .17 (P = 0.05) for the greenhouse test versus plant cane, first stubble, and the mean of plant and first stubble, respectively. The correlation coefficient of bored internodes in the plant cane crop and first ratoon crop was r = .43, (P < 0.01). The low levels of correlation between the reactions of cultivars to sugarcane borer in the greenhouse and field suggested the need for a statistical method to analyze whether in the range of resistance to susceptibility, there was any reaction among cultivars in which the greenhouse and field results agreed. Cluster analysis was considered an appropriate method. The results of a four-cluster analysis is presented in Table 2. Those selections assigned to cluster 1 were considered to have high resistance, cluster 2 contained selections considered to have high susceptibility, and clusters 3 and 4 were considered intermediate with responses dependent on greenhouse or field environment.

Table 3 shows the distribution of clones among quartiles from resistance to susceptibility using data sets from the greenhouse and the two field measurements within the four cluster groups. The cluster group having the most agreement between all three data sets was cluster group 2. This group was comprised of individuals with high susceptibility to borer with only the first ratoon field evaluation assigning clones to a resistant grouping. The next cluster group showing some agreement among data sets was cluster 1. The greenhouse results put all 12 (100%) of these selections into the resistant quartile while the plant cane evaluation assigned 4 (33%) and the first ratoon evaluation assigned 3 (25%) into susceptible quartiles. The remaining 88 clones were divided evenly between cluster groups three and four; their resistance status varied widely within and among the three data sets.

Table 2. Cluster analysis on the proportion of deadhearts from the greenhouse evaluation and of internodes bored in plant cane and first ratoon crops from the field evaluation of 123 selections1.

Cluster means Cluster2 Frequency Greenhouse Plant First Ratoon

1 SAS procedure FASTCLUS (SAS Institute 1987). 2 Pseudo F statistic = 102.88

59

1 2 3 4

(#)

12 23 44 44

(%)

20.1 85.2 85.7 54.1

(%)

12.3 27.5 23.9 15.9

(%)

23.9 35.5 12.2 25.1

Table 3. Distribution of selections among quartiles1 within four cluster groups of Table 2.

1 SAS procedure RANK (SAS Institute 1982).

2 Number of selections (percentage of total in that cluster).

The results of this study suggest that the greenhouse data can be used only as a selection procedure to eliminate or retain the extremes in susceptibility or resistance to sugarcane borer. Further studies must be undertaken to identify all sources of variation operating within the greenhouse if the desired precision needed for selection is to be attained.

REFERENCES

1. Beyer, W. H. (Editor). 1968. Handbook of tables for probability and statistics. Chemical Rubber Co., Cleveland.

2. Breaux, R. D. 1973. Selecting commercial sugarcane varieties from large seedling and clonal population. Proc. ASSCT 2(NS):58-66.

3. Davis, F. M., and W. P. Williams. 1981. Southwestern cornborer: comparison of techniques for infesting corn for plant resistance studies. Jour. Econ. Entomol. 73:704-706

4. Dunckelman, P. H. and B. L. Legendre. 1982. Guide to sugarcane breeding in the temperate zone. USDA, ARS, Agric. Reviews and Manuals, ARS-S-22. 22 pp.

60

2nd Quartile 3rd Quartile 1st Quartile (intermediate (intermediate 4th Quartile

Environment (resistant) resistant) susceptible) (susceptible)

Cluster Group 1 (n=12)

Greenhouse 12(100%)2

Plant cane 6(50%) 2(17%) 3(25%) 1(8%) First ratoon 4(33%) 5(42%) 1(8%) 2(7%)

Cluster Group 2 (n=23)

Greenhouse —- —- 13(57%) 10(43%) Plant cane — — 6(26%) 17(74%) First ratoon —- 3(13%) 5(22%) 15(65%)

Cluster Group 3 (n=44)

Greenhouse —- 1(2%) 22(50%) 21(48%) Plant cane 15(34%) 18(41%) 10(23%) 1(2%) First ratoon 13(30%) 13(30%) 11(25%) 7(15%)

Cluster Group 4 (n=44)

Greenhouse 14(32%) 30(68%) Plant cane 9(20%) 13(30%) 10(23%) 12)27%) First ratoon 13(30%) 10(23%) 14(32%) 7(15%)

5. Everitt, B. 1977. Cluster analysis. Heinemann Educational Books Ltd., London.

6. Pan, Y. S., and S. D. Hensley. 1973. Evaluation of sugarcane seedlings for resistance to the sugarcane borer, Diatraea saccharalis. Environ. Entomol. 2(l):149-54.

7. Ramaiah, B. B., G. N. Rao, and G. H. P. Rao. 1977. Elimination of internodes in sugarcane seed piece. Proc. ISSCT 2:1509-1514.

8. SAS Institute. 1987. SAS/STAT, Guide for personal computers, Version 6. Sas Institute, Cary, N. C.

9. White, W. H. and H. P. Fanguy. 1987. Evaluating potential sugarcane varieties for response to sugarcane borer at the infield stage of selection. Sugar y Azucar 82:27. (Abstract)

10. Wiseman, B. R., F. M. Davis, and J. E. Campbell. 1980. Mechanical infestation devise used in fall armyworm plant resistance programs. Fla. Entomol., 63:425-431.

61

FAMILY PERFORMANCE AT EARLY STAGES OF SELECTION AND FREQUENCY OF SUPERIOR CLONES FROM CROSSES AMONG CANAL POINT CULTIVARS OF SUGARCANE

P.Y.P. Tai and J. D. Miller USDA Sugarcane Field Station

Canal Point, Florida

ABSTRACT The selection scheme of the Canal Point sugarcane (Saccharum spp.) breeding program consists of five

stages, which include Seedling, Stages I, II, III and IV. Visual selection is used in Seedling Stage and Stage I, objective selection is used in Stages III and IV, and a combination of visual and objective selection is in Stage II. From seven Canal Point cultivars used as parents, approximately 12,000 original seedlings were used to evaluate their progeny performance at early stages of selection and to evaluate the frequency of superior clones. Patterns of frequency distribution of stalk diameter of unselected samples varied among progenies with some having fairly symmetrical curves. The stalk diameter of unselected samples ranged from 12 mm to 36 mm in the Seedling Stage and from 18 to 42 mm in Stage I. When stalk diameter was the primary selection criterion for progenies in the Seedling Stage and Stage I, some parents or cross combinations produced higher rates of acceptable clones than did the other parents or cross combinations. Progeny means of measurements of juice quality were less than those of either parent, but most crosses produced some clones that exceeded the performance of either parent. If the clones advanced to the yield test in Stage IV were assumed to be superior ones, the frequency of these clones ranged from one in 1510 to one in 3917 among original seedlings of the seven cultivars used as parents. Choice of both parental clones and cross combinations affect the effectiveness of a sugarcane breeding program.

INTRODUCTION

Evaluation of parent and progeny performance is the most important measure that a sugarcane breeder performs to determine the efficiency of his variety improvement program. Poor parents and/or crosses waste time and funds.

Skinner (12) considered the proven-cross system to be an important method to increase the efficiency of selection by raising the overall quality of original seedling populations. Buzacott (4) suggested that unless crosses are chosen that produce a reasonable percentage of seedlings with a high sucrose content, the likelihood of selecting a suitable commercial type was very remote unless huge populations were handled. Arceneaux (2) recognized that a parent must be judged on its ability to produce seedlings that are superior to existing varieties.

Several methods have been used to evaluate parent and progeny performance in sugarcane. Walker (14) utilized selection percentages from the first three stages as measures of family performance. However, Arceneaux (2) noted that progeny performance, as judged from selection rates in early rounds of screening, was not always a reliable index of value of a parent, but that rates of selection in stages approaching the commercial level were of greater practical significance. George (7) determined that the frequency curves of certain valued agronomic characters, obtained by measuring random samples of seedlings from each cross, enabled the respective merits of each cross to be determined. He also used a grade score to estimate the potential of sugarcane crosses and found that the mean grade was a good guide where difference between the means was large (8). Coleman et al. (5) used breeding rate, calculated by multiplying the percentage of superior plants for each of the evaluated characters together, to predict progeny performance and the environmental adaptation of the cross. Arceneaux et al. (3) used a factor for superior performance (FSP) approach to examine the incidence of superior seedlings from different parental sources and concluded that the high seedling performance would reflect the corresponding level of genetic potential.

A few sugarcane breeders are in favor of using proven parents rather than proven crosses (14). In Barbados, Walker (14) suggested that repeated crosses are yielding diminishing returns both in terms of improved clones and information. In Mauritius, George (7) noted that the work involved in repeated crosses is scarcely worthwhile after the initial production of between 2,000 and 3,000 seedlings. However, the changes in year to year environmental conditions might affect progeny performance. Warner (15) suggested that the superiority of breeding with successive generations of elite hybrids over repeating the crosses between the original parents.

62

He demonstrated that the frequency of elite seedlings increased as the successive generations of elite hybrids were advanced in the Hawaii sugarcane breeding program.

The objective of this study was to evaluate the progeny performance of eleven selected crosses in early and intermediate stages of selection as a means of providing predictive information about the breeding potential of parental clones utilized in a temperate-environment sugarcane breeding program.

MATERIALS AND METHODS

Progenies of 11 crosses produced from seven parental clones were chosen randomly from the regular sugarcane breeding program in 1979. Characteristics of these parental clones are presented in Table 1. Both CP 63-588 and CP 70-1133 had become leading cultivars in Florida after their release. The seedlings were observed through all stages of selection. The selection procedures followed at the USDA, ARS, Sugarcane Field Station, Canal Point, Florida, are outlined in Table 2 (6,9,10). The selection scheme indicates that visual selection criteria are used in Seedling and Stage I and objective criteria in Stage IV. Both visual and objective selection standards are applied to the selection in both Stages II and III.

Table 1. Parentage, date of release and quantitative characters of 7 parental clones of sugarcane used as checks in seedling stage.

1 Released by the U. S. Department of Agriculture, Florida Agricultural Experiment Station and Florida Sugar Cane League, Inc. 2 Released by the U. S. Department of Agriculture, Louisiana Agricultural Experiment Station and American Sugar Cane League of

the USA, Inc.

True seed were planted in flats in the greenhouse in January 1979. Seedlings were kept in the greenhouse until mid-April and then transplanted to the field in two rows 1.5 m apart with 0.3 m between seedlings within a row. Single bud cuttings of the parental clones were planted along with the seedlings of each cross as standards. In early December, stalk number, diameter (at a height of 0.5 m above ground level), and height (as an average of 5 stalks per seedling) were taken on a random sample of 100 seedlings from each cross. One stalk, from each of those 100 seedlings and from the selections from the remaining regular selection was cut 1 m long and advanced as an entry for evaluation in Stage I in December 1979. Each 1-m long stalk was planted as a single plot in rows 1.5 m apart and 0.6 m between plots. Data on three characters (stalk number, stalk diameter, and height) were collected in Stage I from the 100 randomly selected clones of each cross in January 1981. Selection of regular Stage I clones was conducted in September 1980. A 10-stalk seed cane sample was cut from each selected clone in Stage I and used to establish a 2-row 4.6 m long plot in Stage II during October 1980. The number of millable stalks in each plot was determined during August 1981 and a 10-stalk sample from each plot was taken for juice analysis in October. Sugar yield was estimated according to the modified Winter-Carp-Geerligs formula (1). CP 63-588 was used as the standard cultivar in Stage II. By using both visual and objective selection standards, a total of 105 clones (131 clones have been used since 1987) from Stage II test was advanced to the Stage III test. After having been tested for two years (plant cane and first stubble) at four locations across the major cane production areas, 8 to 11 top performing clones were advanced to the Stage IV test at eight locations for three years.

63

Date of Stalk Stalk Variety Parentage release diameter weight Brix Sucrose Purity S/T

CP 63-588 CP 65-357 CP 68-1067 CP 69-1052 CP 70-1133 CP 70-1547 CP 71-1442

CI 54-1910 x CP 57-120 CP 52-68 x CP 53-17 CP 52-68 X CP 57-603 CP 62-374 x CP 56-59 67 P 6 CP 56-63 CP 62-374 x CP 57-526 CP 59-73 x CP 56-63

19681

19732

19751

19791

19771

Exptl. clone Exptl. clone

mm

32.7±4.9 24.8 ± 1.5 31.8±2.9 27.3 ± 1.8 27.8 ± 2.3 29.2 ± 2.8 28.2 ± 2.7

kg

1.9±0.4 1.5±0.2 1.9±0.4 1.4±0.3 1.6 ± 1.3 1.6±0.6 1.6±0.4

°

16.9 ± 1.8 18.7±0.3 17.4 ± 0.9 173 ± 0.8 18.8 ± 1.1 17.7±0.2 16.5± 1.1

% 14.5 ± 1.9 16.1 ± 0.7 14.7 ± 1.6 14.4±2.2 16.7±1.2 14.4 ± 1.4 13.8 ± 1.6

% 85 ± 3 86 ± 3 84 ± 3 82±5 89±2 81 ± 1 83±5

kg

100±15 112 ± 80 102±14 98 ± 20

118 ± 90 97±14 94 ± 14

Mean differences of the various characters among the 11 crosses (Table 3) were tested for significance by Duncan's Multiple Range Test(13).

Table 2. The selection scheme of the Canal Point (CP) sugarcane breeding program.

Original Seedlings 80,000 -1000

25.6 cm spacing apart between seedlings, 1.52 m between rows

One crop (plant cane)

Selection made on 12-month-old seedlings; visual (subjective) selection criteria: vigor, stalk diameter, stalk solidness, stalk height, stalk number erectness, freedom from major disease.

Stage I

3.04 m, single stalk plots; 1.52 m between rows

One crop (plant cane)

5,000 - 12,000 clones

Selection made on 10-month-old plant cane clones; visual (subjective) selection criteria same as seedling stage; permanent CP. number assigned to each selected clone.

Stage II

Two-row, 3.04 m x 4.56 m plot; one location

One crop (plant cane)

1,200 - 1,500 clones

Selection made on 12-month-old plant cane clones; visual selection criteria same as seedling stage; the objective selection criteria include Brix, sucrose, purity, sugar and cane yield, stalk weight.

Stage III

Two-row, 3.04 m x 4.56 m plot; 2 replication; 4 locations;

2 crops (plant cane and 1st stubble)

105 - 131 clones

Selection made on 12-month-old plant cane and first stubble; visual and objective selection criteria same as in Stage II

Stage IV

Four rows, 6.08 m x 10.64 m; 4 replications, 9 locations;

3 crops (plant cane, 1st and 2nd stubble)

8 - 11 clones

Visual and objective selection criteria same as in Stage III. Samples taken at preharvest and harvest; selection standards for release decided by the Florida Sugarcane Variety Committee.

Commercial C. P. Cultivars

64

RESULTS AND DISCUSSION

The percent frequency distribution of millable stalk diameter of unselected samples in both Seedling Stage and Stage I is given in Table 3. Results from both stages indicated that there was a wide range of distribution of stalk diameter in each of the 11 crosses. In most cases, the range of distribution for stalk diameter was narrower in Stage I than in Seedling Stage, but the shapes of the frequency distributions were very similar to each other between stages.

Table 3. Frequency distribution (%) of millable stalk diameter of unselected populations of 11 crosses of sugarcane at Seedling and Stage I.

Cross

CP 65-357 x CP 63-588

CP 65-357 x CP 69-1052

CP 65-357 x CP 70-1133

CP 68-1067 x CP 63-588

CP 68-1067 x CP 69-1052

CP 68-1067 x CP 70-1133

CP 70-1547 x CP 63-588

CP 70-1547 x CP 69-1052

CP 70-1547 x CP 70-1133

CP 71-1442 x CP 63-588

CP 71-1442 x CP 70-1133

Seedling Stage I

Seedling Stage I

Seedling Stage I

Seedling Stage I

Seedling Stage I

Seedling Stage I

Seedling Stage I

Seedling Stage I

Seedling Stage I

Seedling Stage I

Seedling Stage I

12 14 16

1 1

4 7

1

1

1 1

1 1

9

2 9 11

Percent progeny with stalk diameter (mm) greater than or equal

18

4

15

7

3

2

28

17

5 1

22 1

20

15

15 4

21

7

10

3

7

7

18

17 6

17 2

22

23 2

14 2

24 1

13

8

13

15

8

29 2

14 12

18 16

24

26 8

23 14

30 12

18 1

27

18

20 2

18 3

14 4

21 19

17 19

26

20 21

15 21

10 21

22 1

24 4

26 2

17 6

26 4

13 4

29 16

4 21

28

3 26

5 21

5 31

22 9

18 11

18 5

16 9

19 16

3 21

14 19

1 18

30

5 22

1 26

3 20

13 24

4 22

13 14

12 19

5 31

1 35

5 16

1 12

32

2 16

1 9

11

4 26

5 26

5 30

6 24

3 23

15

2 6

7

34

4

5

3

20

21

2 21

2 12

15

11

3 5

3

to: 36

1 1

1 11

9

18

10

6

6

1

38 40 42

7 1

6 2 2

6 4

8

2

2

Mean

22 ab1

2 6 b

21 ab 2 5 b

22 b 2 5 b

24 cd 30 dd

24 cd 30 dd

2 5 d 3 0 d

2 5 d 2 8 c

23 c 2 8 c

22 b 2 8 c

24 cd 2 6 b

19 a 2 3 a

CV(%)

12 11

19 13

14 11

13 10

14 12

13 10

16 11

18 11

17 12

17 15

19 15

1 At separate selection stage, means not sharing a common letter differ significantly at the 5% level of probability according to Dur New Multiple Range Test.

If the culling level for stalk diameter was set at 28 mm in the Seedling Stage, an average of 20.28% of the seedlings would be acceptable. Crosses CP 71-1442 x CP 70-1133 and CP 70-1547 x CP 70-1133 had the lowest percentage of acceptable clones (4% and 2%, respectively) while CP 68-1067 x CP 63-588 and CP 68-1067 x CP 70-1133 had the highest percentage of acceptable clones (40% and 39%, respectively). Among the four female parents examined, progeny of CP 68-1067 had the highest percentage of acceptable seedlings (42.67%), and CP 65-357 had the lowest percentage of acceptable progeny (8.33%). Among the three male parents, CP 63-588 had the highest percentage of acceptable progeny (27.75%), CP 69-1052 was next (20.33%), and CP 70-1133 was the lowest (13.25%).

65

Sugarcane plants in Seedling Stage are generally smaller because they are grown from true seed whereas sugarcane plants in Stage I are larger because they are propagated from vegetative cuttings. Average stalk diameter, from the same unselected progenies tested in Stage I, was about 6 mm larger than that in the Seedling Stage. Therefore, the culling level for the stalk diameter was set at 34 mm for Stage I. The average percentage of acceptable clones was 20.85%, which was nearly the same level of acceptability as in the Seedling Stage (Table 3). CP 68-1067 x CP 70-1133 had the highest percentage of acceptable progeny (40%) and CP 71-1442 x CP 70-1133 had the lowest percentage of acceptable progeny (3%) in Stage I. Progeny of CP 68-1067 had the highest and that of CP 65-357 had the lowest percentage of acceptable clones with a culling level for stalk diameter at 34 mm, as was the case in the seedling population. Both male parents, CP 63-588 and CP 69-1052, had similar numbers of acceptable progenies (21.25% and 22.67%, respectively) in Stage I. These results indicated that the frequency distribution patterns for stalk diameter in both the Seedling Stage and Stage I of unselected samples were different among crosses and repeatable among selection stages. George (7) reported that the distribution curves for yield showed that differences due to genotype were consistent and indicative of the selection potential of each cross. The means and coefficients of variation (CV %) for stalk diameter of the unselected progenies of 11 crosses in both Seedling Stage and Stage I are presented in Table 3. Progeny of crosses with either CP 68-1067 or CP 70-1547 as a female parent produced larger diameter stalks than did the other two female parents (CP 65-357 and CP 71-1442) in both stages. Most crosses had a slightly larger CV % for stalk diameter in the Seedling Stage than in Stage I.

Stalk diameter has been reported to be a better selection criterion because its repeatability in early stages of selection is higher than that of either stalk number or Brix (9). This trait was chosen for further examination of the progeny performance. The percentage of acceptable clones (stalk diameter > 28 mm in Seedling Stage and stalk diameter > 34 mm in Stage I) and the actual selection rate of the eleven crosses are given in Table 4. The selection rate was strongly dependent on the parents. Progenies from two female parents, CP 68-1067 and CP 70-1547, gave a higher selection rate in all three stages than did the other two female parents, CP 65-357 and CP 71-1442. If the selection rate in these two early selection stages was used as the measurement of combining ability (11), the results indicated that CP 68-1067 as female had the best general combining ability. Five crosses, CP 68-1067 x CP 65-588, CP 68-1067 x CP 69-1052, CP 68-1067 x CP 70-1133, CP 70-1547 x CP 63-588 and CP 70-1547 x CP 69-1052, express very good specific combining ability when the selection rates was used as a measure of their progeny performance.

Table 4. Percentage of acceptable clones (stalk diameter >_ 28 mm in Seedling Stage and >_ 34 mm in Stage I) and actual selection rate.

1 Acceptable seedlings were selections made based on the measurement of stalk diameter and the actual selection rate was obtained under a combination of visual (subjective) selection criteria.

The correlations between the percent acceptable clones based on the measurement of stalk diameter alone in both Seedling Stage and Stage I and the actual selection rate based on the visual (subjective) selection criteria alone in Seedling and Stage I or based on both visual and objective selection standards in Stage II and the subsequent stages of selections are shown in Table 5. The correlation of percent acceptable clones between

66

Female

CP 65-357

CP 68-1067

CP 70-1547

CP 71-1442

Avg.

Seedling Stage

Accept. Seedlings Act. selection

Accept, seedlings Act. selections

Acept. seedlings Act. selections

Accept, seedlings Act. selections

Accept, seedlings Act. selections

Male C P 6 3 - CP69-588 1052

10.00 7.80

40.00 18.79

37.00 19.15

24.00 8.94

27.80 13.80

7.00 9.75

27.00 21.49

27.00 9.95

(20.30) (13.70)

CP70-1133

8.00 10.52

39.00 21.17

4.00 18.14

2.00 7.18

13.20 14.10

Avg.

8.30 9.36

35.30 20.48

22.70 15.75

(13.00) (8.06)

Stage 1

Accept, clones Act. selections

Accept, clones Act. selections

Accept, clones Act. selections

Accept, clones Act. selections

Acept. clones Act. selections

C P 6 3 -588

5.00 1.26

39.00 3.98

30.00 3.19

11.00 0.50

21.20 2.23

Male C P 6 9 -1052

5.00 0.97

40.00 2.69

23.00 1.85

(22.70) (1.84)

CP70-1133

4.00 1.63

49.00 5.87

19.00 2.26

3.00 1.22

18.80 3.25

Avg.

4.70 1.26

42.70 4.18

24.00 2.43

(7.00) (0.86)

the Seedling Stage (the culling level for stalk diameter at 28 mm) and Stage I (the culling level for stalk diameter at 34 mm) was highly significant (r = 0.850). The percent acceptable clones in the Seedling Stage was significantly correlated with the actual selection rate in Stages I and II (r = 0.679 and 0.770, respectively), but it was not significantly correlated with the actual selection rate in the Seedling Stage (r = 0.585). The percent acceptable clones in Stage I at the culling level of 34 mm for the stalk diameter was highly correlated with the actual selection rate in all three stages.

Table 5. Correlation coefficients for percent (%) acceptable clones based on stalk diameter and actual selection rate in the early selection stages of sugarcane variety developemnt program.

NS = Non-significant at the 5% level of probability, *, **, Significant at the 5% and 1% levels of probability, respectively.

The seedlings that were advanced to Stage IV of the regular sugarcane breeding program were assumed to be elite clones as shown in Table 6. The average selection rate was one in 2,300 seedlings over all crosses in the program to reach Stage IV test whereas the selection rate was one in 1,200 seedlings from crosses of the selected parental clones. During the process of intense selection, these selected parents might have accumulated more elite genes than did the other parents as reported by Warner (15). In the 1979 seedling program, the crosses made from these seven parents produced approximately one elite seedling in 2,000 seedlings tested. Based on the frequency of elite seedlings from these crosses, there should be 30,000 to 40,000 seedlings planted to produce one cultivar for commercial production.

In the Canal Point (C.P.) cooperative sugarcane cultivar development program in Florida, both visual and objective selection standards were applied in the evaluation of the Stage II test before the superior clones were selected for advancement to the Stage III test. The means and ranges for stalk weight and quantitative measures of juice quality of selected clones from eleven crosses tested in Stage II are summarized in Table 7. Progeny means of all characters to measure juice quality were less than those of their parents, but most crosses produced at least some progenies that exceeded the range of parental characteristics. Transgressive segregation provide sugarcane breeders the opportunity to obtain clones which are superior to the parental clones. Sugarcane is a vegetatively propagated crop. Once the superior segregants have been detected, the genotypic characteristics of the clones can be fixed. Among the eleven crosses, CP 65-357 x CP 69-1052 and CP 71-1442 x CP 70-1133 did not produce any clones with transgressive recombinations for Brix, % sucrose, and S/T. None of their progenies tested in Stage II was selected for advancement to Stage III. One cross, CP 70-1547 x CP 70-1133, had more than 1000 seedlings in the initial stage. There were 198 clones obtained after the first round of selection (clones advanced from Seedling to Stage I) and 21 clones obtained after the second round of selection (clones advanced from Stage I to Stage II), but none of these 21 clones tested in Stage II were selected for planting in Stage III because of their low sucrose content and sugar yield. Therefore, parents, that can produce progeny with a high frequency of transgressive recombination for both cane yield and juice quality should provide the best opportunity for sugarcane breeders to select clones superior to parents. Although transgressive recombinations allow selection of superior clones, their performance can be caused by environmental influence rather than

67

improved genetic potential. Therefore, clones must be evaluated in different years and at different locations in replicated tests so that only superior clones, with improved genetic potential, will be released for commercial production.

Table 6. The total number of seedlings and the selection rate (%) in seedlings, Stage I, and Stage III for the CP 80-series.

Table 7. Means and ranges of the eleven crosses for Stalk weight, Brix, Sucrose, Purity and S/T in Stage II.

* = Some clones exceeding the ranges of either parent.

The percent frequency distribution for stalk weight, Brix, % sucrose, and S/T of both selected and total clones of Stage II are shown in Figure 1. The results showed that means for both stalk weight and juice quality of the selected clones were moved toward higher values under selection pressure. The clones in Stage II had mean values of 1.87, 17.35, 13.81 and 89.88 for stalk weight, Brix, % sucrose and S/T, respectively, whereas the clones selected for advancement to Stage III had mean values of 2.02, 18.54, 15.87 and 108.43, for the same characters. The frequency distribution of the total population of Stage II appeared to be fairly normal for all

68

four characters. The selected clones had higher means but smaller CV(%) for those four characters than did the entire population in Stage II.

Figure 1. The percent frequency distribution for stalk weight, Brix, percent sucrose, and S/T of both selected and total clones of Stage II.

The evaluation of both parental and progeny performances indicated that the selection rate was strongly dependent on parentage and cross combination. Most of the commercial varieties used as parental clones belong to the complex hybrids of Saccharum spp. Clones that produce a high frequency of transgressive recombination for many agronomic traits in various crosses are highly desirable.

REFERENCES

1. Arceneaux, G. 1935. A simplified method of making theoretical sugar yield calculations in accordance with the Winter-Carp-Geerligs formula.Int. Sugar J. 37:264-265.

2. Arceneaux, G. 1968. Breeding sugarcane varieties for the Northern Caribbean. Proc. ISSCT 13:1034-1046.

3. Arceneaux, G., J. F. Van Breenman and J. O. Despradel. 1986. A new approach in sugarcane breeding: comparative study of progenies for incidence of superior seedlings. Sugar Cane, 1986, No. 1, pp. 7-10.

4. Buzacott, J. H. 1956. Breeding for high sugar. Proc. ISSCT 9:553-560.

69

5. Coleman, O. H., J. L. Dean, and D. M. Broadhead. 1963. Evaluation of sugarcane crosses. Proc. ISSCT 14:483-488.

6. Falgout, R. N., N. I. James and E. R. Rice. 1968. Development of new sugarcane varieties for Florida. Sug Bull. 46(23):7-12.

7. George, E. F. 1959. Effect of the environment on components of yield in seedlings from five Saccharum crosses. Proc. ISSCT 9:755-765.

8. George, E. F. 1962. Applications of a grade score in determining the potential of sugarcane crosses. Proc. ISSCT 11:498-504.

9. James, N. I., and J. D. Miller. 1971. Selection in two seedling crops of four sugarcane progenies. Crop Sci. 11:245-248.

10. Miller, J. D. 1971. USDA sugarcane selection program in Florida. Proc. ASSCT I(NS):145-148.

11. Miller, J. D. 1977. Combining ability and yield component analyses in a five-parent diallel cross in sugarcane. Crop Sci. 17:545-547.

12. Skinner, J. C. 1971. Selection in sugarcane: A review. Proc. ISSCT 14:149-162.

13. Steel, R. G. D., and J. H. Torrie. 1960. Principles and Procedures of Statistics. McGraw-Hill Book Co., Inc., New York.

14. Walker, D. I. T. 1963. Family performance at early selection stages as a guide to the breeding program. Proc. ISSCT 11:469-483.

15. Warner, J. N. 1953. the evolution of a philosophy on sugarcane cane breeding in Hawaii. Hawaiian Planters' Record 54:139-162.

70

YIELD EFFECTS OF SUGARCANE SMUT INFECTION IN FLORIDA

B. Glaz USDA-ARS Sugarcane Field Station, Canal Point, FL

M.F. Ulloa New Hope Sugar Cooperative, Pahokee, FL

and R. Parrado

F & W Farms, Lake Harbor, FL

ABSTRACT

Sugarcane smut, Ustilago scitaminea H. Syd., has been an important disease in Florida and many other regions where sugarcane is grown. The primary objective of this study was to quantify yield losses due to smut under field conditions in Florida from the plant-cane through the second-ratoon crop. An additional objective was to compare the yields of four cultivars, CP 72-1210, CP 73-1547, CP 75-1091, and CP 57-603. These cultivars were chosen to represent smut susceptibilities of resistant, moderately susceptible, susceptible, and highly susceptible, respectively. To obtain different smut levels within each cultivar, seed pieces were either treated with hot water and fungicide or inoculated by immersion in a suspension of viable smut teliospores. Smut levels were quantified by counting whips (sori) and by removing whips and weighing them. There was less variability when determining yield losses by number of whips rather than by whip mass. Actual levels of smut produced were not as expected for all treatments. Both cane and sugar yields were reduced linearly by increased smut incidence. Removing smut whips from infected plants in May and June of each growing season did not alter the effects of the disease on final yield characteristics. The highest infection level attained, 6,265 whips ha-1, reduced sugar yields by 3.85 metric tons ha-1. When quantified by mass, the highest infection level of 1.15 kg ha-1 caused a sugar yield reduction of 2.69 metric tons ha-1. This experiment was conducted on a "warm-land" sugarcane location (land close to the moderating temperature effects of Lake Okeechobee) in Florida. In addition, all 3 harvests (plant-cane through second-ratoon) were late in the harvest season (February or March). Under these conditions, CP 72-1210 had significantly higher sugar yields than all other cultivars, regardless of smut treatment.

INTRODUCTION

Sugarcane smut, Uslilago scitaminea H. Syd., was first reported in Florida in 1978 by Todd (10). Soon after the arrival of smut in Florida, several commercial cultivars were phased out of production due to their high susceptibility. Since that time, Florida sugarcane breeders have incorporated successful strategies in their breeding programs to select cultivars resistant to smut. The high levels of resistance of current commercial cultivars have now relegated sugarcane smut to a minor problem in Florida.

Initial reports by Holder (4) of cane yield losses of 32% from smut infection levels of 50-60% infected stools provided a sound rationale for rigorous selection strategies. In subsequent reports, Glaz et al. (3) and Irey (6) reported that yield losses were considerably less than those reported by Holder. In these reports, except for one commercial cultivar studied by Irey, yield losses of highly susceptible noncommercial cultivars were reported. In other sugarcane producing regions, Olufolaji (7) in Nigeria and Hoy et al. (5) in Louisiana reported that yield losses due to smut were similar to those reported by Holder (4). The major objective of this study was to determine the effects of sugarcane smut on sugar concentration, cane yield, and sugar yield during a three-crop cycle for four important sugarcane cultivars. In addition, this study was designed to compare the reaction to hot-water treatment of seed pieces, the production capabilities, and the effects of roguing of smut whips from infected stalks among the four cultivars.

MATERIALS AND METHODS

The experiment was planted 23-24 Sept. 1982 at Rutledge Farms on a Florida organic soil classified as Torry muck by Snyder et al. (8). Due to poor germination on plots where seed pieces of two cultivars, 'CP 57-603' and 'CP 75-1091', were hot-water treated, the skips in these plots were replanted on 18 Jan. 1982. The

71

experiment was harvested three times: the plant-cane crop was harvested 18 Feb. 1984, the first-ratoon crop was harvested 12-13 Feb. 1985, and the second-ratoon crop was harvested 4 March 1986.

The experiment was planted in a split-plot arrangement of a randomized-complete block design with four replications. Main plots were cultivars and sub plots were disease treatments. The four cultivars and their expected smut susceptibilities were 'CP 72-1210'-resistant, 'CP 73-1547'-moderately susceptible, 'CP 75-1091'--susceptible, and 'CP 57-603'~highly susceptible. The three disease treatments were hot-water and fungicide treatments of seed pieces, smut inoculation of seed pieces, and smut inoculation of seed pieces with subsequent roguing of smut whips (sori) from infected plants. The hot-water treatment was accomplished by immersing seed pieces in water maintained at 52° C with 500 ppm difolatan1 for 45 minutes. Smut inoculation was accomplished by immersing seed pieces for 10 minutes in 284 liters of water at ambient temperature mixed with smut teliospores from 125 smut whips. Seed pieces in both hot-water and inoculated treatments were 44 cm in length. In obtaining seed pieces, the lower three buds from each full length stalk were discarded. All cane was planted within 2 hr of having received hot-water or inoculation treatments. Smut whips were rogued twice (May and July) per crop in the plots that received the roguing treatment. The whips that were rogued were collected in paper bags, dried, and weighed. Smut whips in all plots were counted from May through July of each crop season. Whips were marked with a ribbon at each counting to avoid counting them a second time.

Plots were four rows wide and 9.1 m long. Row and alley spaces were 1.5 m. There were no border rows between plots although there were border rows around the edges of the experiment. Stalks were counted in all rows of all plots on 14 July 1983 in the plant-cane crop, 24 and 31 May 1984 in the first-ratoon crop, and 26 August 1985 in the second-ratoon crop. The cane was cut by hand and weighed with a tractor-mounted weighing device at each harvest to determine cane yield (CY) measured as metric tons cane ha'1. Ten full length, non damaged stalks were randomly selected from each plot for milling and crusher juice analysis. All values of sugar concentration (SC) were theoretically calculated from the Brix and polarity of these samples using Arceneaux's (1) modification of the Winter-Carp-Geerligs formula and were reported as kg sugar per metric ton of cane. Sugar yield (SY) was measured as metric tons of sugar ha-1 and was equal to CY x SC/1000.

All data were analyzed by analysis of variance as a split plot in time and space as described by Steel and Torrie (9). For the ratoon-crop data of CP 57-603 and CP 75-1091, SC, CY, and SY were regressed on number of whips ha"1 and on weight of whips ha-1.

RESULTS AND DISCUSSION

Heavy rains, beginning within 1 hr after the experiment was planted, kept the field flooded for the next 7 days. Subsequently, there were large, unexpected differences in germination due to treatment. The seed pieces of CP 75-1091 and CP 57-603 that had been treated with hot water germinated poorly. These plots were replanted. However, the material that was replanted germinated at levels that were similar to those of the original planting. Thus, stalk counts were low for these treatments (Table 1). The field was not excessively wet when the two cultivars were replanted. Therefore, it was probably the hot-water treatment rather than the flooding that caused the poor germination in CP 75-1091 and CP 57-603.

The original intent of this experiment was to grow four cultivars varying in smut susceptibility from resistant to highly susceptible and to further subdivide levels of smut infection (particularly in the plant-cane crop) by treating seed pieces with hot water (to reduce smut incidence) and inoculating seed pieces with smut teliospores (to increase smut incidence). Although the experiment was planted in an area known to have a high natural incidence of smut, there was no smut present on any treatments in the plant-cane crop. Smut was present in the ratoon crops, but its levels did not correspond as expected to the hot-water and inoculation treatments (Table 2). Also, CP 72-1210 and CP 73-1547 had no smut throughout the 3-year study. Thus the smut yield-loss estimates discussed here are based only on first- and second-ratoon data of CP 75-1091 and CP 57-603.

1Mention of trade name or proprietary product does not imply or constitute endorsement or recommendation by the USDA.

72

Table 1. Stalk counts of four cultivars subjected to three smut treatments from the plant-cane through the second-ratoon crop.

1 PC = plant cane, 1R = first ratoon, 2R = second ratoon

2 Errors a - d correspond to the errors a - d described by Steel and Torrie (9).

Table 2. Smut levels in hot-water treated and inoculated CP 57-603 and CP 75-1091.

Most of the attention of this study was devoted to the effects of the various treatments on sugar concentration (SC), cane yield (CY), and sugar yield (SY). Yield loss estimates are reported in relation to number of smut whips ha-1. Yield losses were also calculated in relation to weight of smut whips ha-1, but these

73

estimates are not reported because they were similar to but not as precise as those in relation to number of whips ha-1.

Sugar Concentration. The disease treatments had little to no effect on SC as indicated by their low F values (Table 3). However, Comparison 19 of Table 3 indicates that in the plant-cane crop, CP 72-1210 treated with hot water had a lower SC than inoculated CP 72-1210, whereas these results were opposite with CP 75-1091. The treatment effects reversed in the second-ratoon crop when hot-water treatment gave a higher SC for CP 72-1210 and smut inoculation resulted in a higher SC for CP 75-1091 (Table 4). CP 72-1210 had no smut throughout the study so these treatment effects on it cannot be explained as effects of smut. However, for CP 75-1091, the level of smut was considerably lower in the hot-water treated as opposed to the inoculated plots in the second-ratoon crop (Table 2). Perhaps the increased smut incidence in the inoculated CP 75-1091 in second ratoon caused the increase in SC although this effect was not repeated in other comparisons.

Table 3. F values and their levels of probability for sugar concentration, cane yield, and sugar yield single degree of freedom comparisons.

1 H = hot-water treated seed pieces, I = seed pieces inoculated with smut, IR = seed pieces inoculated with smut and subsequently rogued. 2 Errors a, b, c, and d are reps x cultivars, reps x cultivars x crop, reps within cultivars x crop, and reps within cultivars x disease treatment x crop, respectively.

74

The linear and quadratic regressions of SC on number of whips ha'1 were not significant.

There were some important SC differences due to effects not related to disease treatments. CP 72-1210 had a significantly higher SC than the other three cultivars in all three crops (Comparisons 1-3 and 6-8 Table 3, and Table 4). When only the effect of crop was considered, SC increased significantly from the plant-cane crop to the first-ratoon crop and then dropped significantly in the second-ratoon crop (Comparison 5 Table 3, and Table 4). The low SC in the second-ratoon crop was probably due to below freezing temperatures that occurred shortly before the second-ratoon harvest.

Table 4. Sugar concentrations of four cultivars subjected to three smut treatments from the plant-cane through the second-ratoon crop.

1 PC - plant cane, 1R = first ratoon, 2R = second ratoon.

Cane Yield. Overall, the cane treated with hot water had a significantly lower CY than the cane that was inoculated with smut (Comparison 9 Table 3, and Table 5). However, there were many significant interactions of disease treatment with cultivar and crop. For comparisons of disease treatment and CP 72-1210 with CP 57-603 or CP 73-1547, there were significant interactions with crop (Comparisons 18 and 20 Table 3). CP 72-1210 had a reduced CY in the plant-cane crop, but increased CY's in the ratoon crops due to hot-water treatment of seed pieces (Table 5). Since hot-water treated plots of CP 72-1210 had reduced stalk counts in both the plant-cane and first-ratoon crops, the major cause of the CY differences was probably something other than the reduced germination caused by the hot-water treatment (Table 1). Hot-water treatment caused reduced CY's in the plant-cane and first-ratoon crops and no change in CY compared to smut inoculated cane in the second-ratoon crop for CP 57-603 (Table 5). Since the hot-water treated and inoculated plots of CP 57-603 had similar relative smut incidence in the two ratoon crops, the relative improvement in the CY production of the CP 57-603 hot-water treated plots in second ratoon was not due to decreased incidence of smut. In this case, most of the yield differences among crops was probably related to the reduced germination caused by hot-water treatment of CP 57-603 (Table 1). In the first two crops when effects of hot-water treatment on number of harvestable stalks were highest, the CY's of the hot-water treated plots were lower than those of the inoculated plots. However, by second ratoon when the number of harvestable stalks was similar for both treatments, the CY's were no longer significantly different (Table 1 and Table 5).

75

Table 5. Cane yields of four cultivars subjected to three smut treatments from the plant-cane through the second-ratoon crop.

1 PC = plant cane, 1R = first ratoon, 2R = second ratoon.

Hot-water treatment had no effect on the CY's of CP 73-1547 in the first two crops, but it did cause a small decline in CY in the second-ratoon crop (Table 5). Since there was no smut in any CP 73-1547 treatments, and stalk numbers were not affected by disease treatments with CP 73-1547 (Table 1), it is difficult to explain the cause of this yield loss in the second-ratoon crop. These summaries assumed that lower levels of CY's for hot-water compared to smut inoculated treatments were due to hot-water treatment having caused reduced CY's rather than smut inoculation having caused increased CY's. We know of no evidence that smut inoculation increases CY, but Benda (2) has demonstrated that hot-water treatment of seed pieces can reduce germination in some cultivars.

Linear regression provided the best fitting estimate of CY decline due to increasing incidence of smut infection (Fig. 1). As were the r2 values reported previously by Glaz et al. (3) and Irey (6), the r2 value of this regression was significant but low. Thus, the 95% confidence intervals of means have also been included in Fig. 1. We can be 95% certain that at any level of smut from 0 to 6000 whips ha-1, the mean CY will be within the corresponding interval shown. The equation shows that for every 1000 whips ha-1, CY declined by 4.81 metric tons ha-1. In previous studies, Glaz et al. (3) and Irey (6) reported CY losses of 2.3 and 4.1%, respectively, with 10% infected stalks. In Nigeria, Olufolaji (7) and in Louisiana, Hoy et al. (5) reported losses in CY of 6 to 10% at 10% smut infection levels. Assuming stalk populations of 61,000 ha-1, the present study predicts CY declines of about 30% with 10% infected stalks. One factor that may have caused the yield losses related to smut to be higher than expected was the reduced stalk counts in the hot-water treated CP 57-603 and CP 75-1091 plots. However, these stalk counts were only reduced in the plant-cane and first-ratoon crops (Table 1). Since the regressions describing smut were only calculated from first- and second-ratoon data, it is not expected that the stalk count factor accounted for the large magnitude of difference between this and other studies.

The logical conclusion of the previous information is that under some environmental conditions, smut may cause more yield losses than previously suspected. However, the present study also presented results that conflict with this conclusion. The plots of CP 57-603 that were inoculated and rogued had higher smut levels than the CP 57-603 plots that were inoculated (Table 2). However, these increased smut levels did not cause

76

a decline in CY (Tables 3 and 5). The most prudent overall conclusion is that since there has been success in selecting smut resistant clones in Florida, this overall strategy should be continued.

Figure 1. Linear regression and 95% confidence limits of means of metric tons of cane per hectare on number of smut whips per hectare.

Sugar Yield. As with CY, hot-water treatment of seed pieces caused a significant reduction in SY (Comparison 9 Table 3), and as with CY, this effect was not similar for all cultivars (Comparisons 11-13 Table 3). Hot-water treatment had no effect on the SY's of CP 72-1210 or CP 73-1547, whereas hot-water treatment caused significant reductions in SY of CP 75-1091 and CP 57-603 (Table 6). For both cultivars with reduced SY's, the cause was probably the lower germination after hot-water treatment (Table 1). For CP 57-603 a second probable cause was the increased smut incidence in the plots that were treated with hot water (Table 2). As with CY, linear regression provided the best fit to describe SY decline with increasing smut incidence (Fig.2). Sugar yield declined by 0.62 metric tons ha-1 with every increase of 1000 smut whips ha-1.

77

Figure 2. Linear regression and 95% confidence limits of means of metric tons of sugar per hectare on number of smut whips per hectare.

78

Table 6. Sugar yields of four cultivars subjected to three smut treatments from theplant-cane through the second-ratoon crop.

1 PC = plant cane, 1 R = first ratoon, 2R = second ratoon.

In treatment comparisons not related directly to disease treatments, the cultivar with the highest SY across all treatments was CP 72-1210 (Comparisons 1-3 Table 3, and Table 6). Overall, SY declined linearly (Comparison 4 Table 3) from 17.4 metric tons ha-1 in the plant-cane crop to 8.2 metric tons ha-1 in the second-ratoon crop.

There were no significant differences between inoculated cane that was rogued and not rogued for SC, CY, or SY (Comparison i0 Table 3). Also, there were no significant interactions of this comparison with cultivars or crops (Comparisons 14-17 and 21-22 Table 3). Therefore, roguing did not affect SC, CY, or SY, neither in the year in which the cane was rogued, nor in the second-ratoon crop after having been rogued in the first-ratoon crop.

REFERENCES

1. Arceneaux, G. 1935. A simplified method of making theoretical sugar yield calculations. In accordance with Winter-Carp-Geerligs formula. Int. Sugar Jour. 37:264-265.

2. Benda, G.TA. 1988. Hot-water cure of sugarcane mosaic in sugarcane stalks in Louisiana. Jour. ASSCT 8:56-61.

3. Glaz, B., J.L. Dean, and J.D. Miller. 1985. Yield effects of various levels of sugarcane smut infection in Florida. Jour. ASSCT 4:50-53.

4. Holder, D.G. 1983. Influence of smut on production in highly susceptible varieties of sugarcane. Jour. ASSCT 2:29-31.

5. Hoy, J.W., CA. Hollier, D.B. Fontenot, and L.B. Grelen. 1986. Incidence of sugarcane smut in Louisiana and its effect on yield. Plant Disease 70:59-60.

79

6. Irey, Michael S. 1986. Yield losses in sugarcane variety CL 65-260 due to sugarcane smut in Florida. Jour. ASSCT 6:32-36.

7. Olufolaji, D.B. 1987. Yield loss through cane smut in Nigeria. Sugarcane 2:6-7.

8. Snyder, G.H., H.W. Burdine, J.R. Crockett, and others. 1978. Water table management for organic soil conservation and crop production in the Florida Everglades. Univ. Fla. Inst. Food Agr. Sci. Tech. Bull. 801, 22 p. 9.

9. Steel, Robert G.D. and James H. Torrie. 1960. Principles and Procedures of Statistics. p. 247-249. McGraw-Hill Book Company, Inc. New York.

10. Todd, E.H. 1978. Sugarcane smut in Florida. Sugar J. 41(3):23.

80

NUTRITIONAL STATUS SURVEY OF SUGARCANE IN TEXAS1

J. R. Thomas and N. Rozeff Soil Scientist, USDA, ARS, Weslaco, TX 78596 and

Agriculturist, Rio Grande Valley Sugar Growers, Inc., Santa Rosa, TX 78592

ABSTRACT

Nutrient imbalance may contribute to the low average sugarcane tonnage in the Lower Rio Grande Valley (LRGV). A survey was conducted to determine the extent of reported K deficiency and to evaluate the nutrient status of sugarcane in the LRGV. The diagnostic recommendation integrated system (DRIS) and critical nutrient levels were used to diagnose the nutrient status of the sugarcane fields. Based on critical levels, N was the primary nutrient limiting cane growth. Nitrogen was deficient in 38% of the 3-month-old cane surveyed. Phosphorous and potassium deficiencies were found in 11 and 8% of the fields, respectively. Low leaf P and K levels may reflect the effects of soil moisture and N on P and K availability and uptake. Adjustment of the low leaf P and K concentrations for the effects of N raised the P and K values into their normal range. DRIS analysis suggested that 34% of the 3-month-old cane surveyed required K to achieve nutrient balance. Nitrogen and phosphorous were required in 49 and 14% of the surveyed fields, respectively.

INTRODUCTION

The low average per acre sugarcane tonnage in the Lower Rio Grande Valley (LRGV) may be associated with nutritional disorders. Nitrogen (N) and iron deficiencies in ratoon crops have occurred on all cultivars and soil types (11, 12, 15). Deficiency symptoms for phosphorous (P) has not been observed, but leaf P concentrations of some cane fields were near the critical level. However, no yield response to P fertilizer has been obtained (16). Leaf K concentrations below the critical level have been reported (16). However, potassium (K), calcium (Ca), and magnesium (Mg) levels in LRGV soils are believed to be adequate for full growth. Hipp (8) reported that clay minerals in the LRGV soils had a large capacity to supply K to plants and to replenish the soil K used by plants.

It is generally recognized that plant growth depends on the concentration or intensity of different essential nutrients above a critical level and on the balances between them. Foliar analysis is used in many sugarcane growing areas to diagnose possible nutrient deficiencies and imbalances (1, 5, 6). In this study, the nutrient composition of leaves from low and high yielding sugarcane fields were monitored throughout the growing seasons to determine the extent of a possible K deficiency and to evaluate the nutrient status of sugarcane in the LRGV. The relationships between nutrient concentration, nutrient ratios, cane yield and sugar content were evaluated.

MATERIALS AND METHODS

Critical nutrient levels (6) and ratios established in other areas and confirmed or modified by fertilizer studies in LRGV were used to diagnose the nutrient status of the sugarcane fields. Nitrogen sufficiency curves (14) for cultivars NCo 310, CP 52-68, and CP 65-357 suggested that leaf N levels for 3-month-old cane should be 2.18, 2.10, and 2.00%, respectively. Corresponding levels for 6-month-old cane were 1.68, 1.73, and 1.66%. Normal concentrations from the literature (6) and used for all cane ages were P (0.24 to 0.18%), K (1.35 to 1.02%), Ca (0.16-0.20%) and Mg (0.08-0.19%). As N may affect the uptake of P and K, the leaf concentrations of P and K were adjusted for N (X) by the relationships:

1Contribution of the Conservation and Production Systems Research Unit, ARS, USDA, Weslaco, TX in cooperation with Rio Grande Valley Sugar Growers, Inc., Santa Rosa, TX

81

adj PN = P + 0.1172 (X - x) and adj KN = K + 0.545 (X - x)

where x is the normal leaf N concentrations at a given age. The Diagnosis and Recommendations Integrated System (DRIS) was used to evaluate the balance among the various nutrients (1, 2). DRIS diagnostic norms established for sugarcane in Florida and South Africa were based on the mineral compositions of the top visible dewlap (TVD) leaf laminae. However, nutrient norms established for the TVD leaves may not apply when other leaves are selected. In this study the normal nutrient ratios were based on the mineral composition of leaves 3 through 6, counting the spindle leaf as No. 1, from cane 3 to 6 months of age. The leaves were collected from sugarcane that produced more than 45 tons/acre. The Texas norms are slightly different from the norms used in Florida (Table 1). Correlations between the N, P and K concentration of the TVD and the 3 through 6 leaves were highly significant. The coefficient of determination (r2 x 100) were 85, 89 and 84%, respectively, for N, P and K. The DRIS diagnostic norms were assumed to be constant throughout the growing season (10).

Table 1. Normal nutritional ratios used.

1 3 through 6-month-old cane, leaves no. 3 through 6. 2 TVD leaf (6).

The availability of soil water is a major yield-determining factor not explicitly evaluated in the DRIS diagnosis. The effects of moisture stress on the plants nutrient composition were minimized by collecting leaf samples within a few days of an irrigation or rainfall event. However, as the N, P and K concentrations were significantly related to the plant's moisture status, the DRIS approach was modified to evaluate the nutrient-moisture relationship. Normal nutrient-moisture ratios between leaf N or P concentrations and the leaf sheath moisture (M) levels were established. The leaves and sheaths were collected from N fertilized and irrigated sugarcane that produced more than 45 tons/acre. The normal N/M and M/P ratios were the means of 910 and 721 ratios, respectively.

Since sheath moisture changed throughout the growing season, the relation: % M = 86.95 - 0.26X, r = 0.89, between sheath moisture (M) and cane age (X) expressed in terms of climatological weeks was used to calculate the normal sheath moisture level for a given age. March 1-7, inclusive, is climatological week number one and closely approximates the dates of growth initiation. In week 12 (3-month-old cane) the normal sheath moisture level was 83.8%. Corresponding values for 4 and 6-month-old cane were 82.7 and 81.7%.

82

Thirty-seven commercial sugarcane fields were selected to represent fields with high and low production records and different soil types (Table 2). The sampled fields occupied 4.8% of the sugarcane acreage in the LRGV. Five fields contained more than one cultivar. One field had two soil types.

Table 2. Soils included in study.

Leaves and sheaths numbered 3 through 6 (spindle leaf No. 1) were collected from five stalks at each sampling site at 3, 4 and 6 months of age. The sheaths were separated from the leaves at the dewlap and dried at 70°C for moisture determination. The middle third of the leaf blades with midribs removed were dried (70°C) ground to pass a 2 mm sieve, and analyzed for total N (4), P(3), K, Ca, and Mg by atomic absorption method. Data on sugarcane tonnage, sugar yields and juice quality were obtained from commercial field records (Cowley Sugar House). The vegetative growth index, expressed on the green sheath weight, was used as an indicator of cane growth. Correlation and regression analysis were used to evaluate the relationships between yields, cane growth, juice quality and leaf mineral composition.

Solar radiation data were obtained from measured values at the USDA laboratory in Weslaco, Texas.

RESULTS AND DISCUSSION

Production records suggested that half of the selected fields should produce at least 45 tons of cane per acre; however, the actual yields were lower than expected. Fields with high or low yield records averaged 37.9 and 25.6 tons/acre, respectively. Freezing temperatures during January 1982 affected the plant population of the 1982-83 crop and the low average daily radiation of 381 gm cal/cm2/day during the 245 day growth period, March 1 through October 31, 1982, decreased plant growth. The average daily radiation for the previous 20 years was 456 gm cal/cm2/day for the same period. The low average daily radiation reflects the effect of volcanic ash from the El Chichon volcanic eruption (9).

Yields of cane and sugar listed by cultivar, crop cycle, and soil types are given in Table 3. Variation in yields and juice quality were high. Cane and sugar yields ranged from 11.5 to 48.7 and 0.92 to 4.93 tons/acre, respectively. Pol ranged from 9.5 to 13.3% and juice purity from 75.2 to 84.5%. The cane fields varied in cycle from plant through the 9th ratoon. Eight percent of the fields were in plant cane, 51.5% were 1st, 2nd and 3rd ratoon crops, 21.6% were 4 through 8 ratoons and 18.9% were 9th ratoon crops. No 5th ratoon crops were in the survey. Plant cane which was expected to produce the highest tonnage averaged 27.5 tons/acre whereas 1st and 2nd ratoon crop yields averaged 31.9 tons/acre. The lowest yield (0.6 tons/acre) was produced by an unfertilized and nonirrigated 9th ratoon crop of NCo 310 on Raymondville clay loam. However, on fertilized and irrigated fields, 9th ratoon crop yields averaged 27.6 tons/acre. No significant relationships were found between cane yields and cultivars, crop cycle or soil types.

83

Table 3. Soil type, cultivar, sugarcane cycle, yields and juice quality.

Cycle Plant cane 1 Ratoons 2-10

84

Table 4. Range and means of leaf mineral content and nutrient ratios at three cane ages.

Cane fields in need of irrigation or fertilization were easily identified by low sheath moisture and leaf N levels. The range and means of the leaf mineral concentration and nutrient ratios are presented in Table 4. The nutrient concentration and ratios among N, P, K, Ca, and Mg in the leaf tissues changed as the cane matured. The leaf N, P, and K concentrations in the 3-month-old cane were lower than normal in 38, 11, and 8%, respectively, of the sampled fields (Table 5). However, when the leaf P and K values were adjusted for the effects of N, none of the fields were deficient in P or K. The low P and K concentrations were associated with N deficient cane. The Ca and Mg values were higher than those considered normal for adequate nutrition. Some fields were fertilized after May 3rd; this reduced the percentage of fields with N and P deficient cane. However, the leaf N concentrations in all fields of 6-month-old cane were below normal, approximately 2 months before harvest of the early cultivars.

Low sheath moisture levels probably reflected the low N status of the crops and low soil water availability. Clements (5) states that "any element in deficient supply lowers tissue moisture" and Thomas et. al. (13) found that sheath moisture was significantly affected by the irrigation regime and row spacing. Significant correlation coefficient r = 0.69 and r = 0.70 for 3 and 4-month-old cane, respectively, suggested that leaf N and sheath moisture were interdependent.

85

Growth of sugarcane, as indicated by the vegetative growth index (VGI), was significantly influenced by the N, K and moisture status of the plant. The relation between VGI (Y), leaf N and K concentration and sheath moisture (M) of 3-month-old cane is given by the regression equation.

Y = - 192.24 - 24.93N + 16.45K + 3.10M (r = 0.65).

The VGI is associated with stem girth (5) and is generally used to monitor the effects of soil moisture or N fertilization on cane growth.

Table 5. Evaluation of nutrient status of sugarcane in Texas based on nutritional concentrations in laminae of leaves 3, 4, 5, and 6.

Cane yields were significantly correlated (Table 6) with the moisture and N status of the crop at 3, 4, and 6 month of growth, but the degree of correlation was low. Variation in leaf N and sheath moisture levels of the 3-month-old cane accounted for 12 and 17.5%, respectively, of the yield variability. Nonsignificant partial correlation coefficients ryn.m = 0.08 and rym.n = 0.26 show that neither N nor sheath moisture influenced yields independently of each other. Stepwise regression analysis indicated that cane yields were significantly related to the N, P, Mg and sheath moisture content of 3-month-old cane. Variation in the percentages of these three nutrients and sheaths moisture accounted for 43% of the yield variability. The relationship between yields and leaf K percentage of 6-month-old cane may reflect interaction between N and K or K and sheath moisture. However, partial correlation coefficients between yields and K at the same N level (ryk.n = 0.43) and at the same sheath moisture level (ryk.m = 0.37) were significant indicating that K influenced yields independent of N or moisture. Nonsignificant partial correlation coefficients (rnk.m = 0.23) between N and K at the same sheath moisture level implied that the N K correlation reflected the influence of moisture availability rather than N on K uptake. Figure 1 shows that sheath moisture influences the leaf K concentration.

DRIS (2) indices based on N/P, N/K and K/P ratios suggested that N, P, or K was the most required nutrient in 49, 14, and 34%, respectively, of the 3-month-old cane surveyed (Table 7). A few fields required both N and K. Whereas, DRIS indices based on ratios involving five elements implied that N, P, K, Ca or Mg were required in 25, 5, 27, 15, and 28%, respectively, of the fields to achieve nutrient balance. DRIS indices based on 10 ratios are reportedly more accurate than indices calculated from 3 ratios (7). Lack of yield responses to Mg or K (16) by sugarcane on LRGV soils prevents the determination of the accuracy of the Mg and K diagnosis.

DRIS indices range from negative to positive values depending on whether the nutrient was relatively deficient or excessive with respect to the other nutrients (2). The degree of nutritional balance decreases as the

86

absolute sum of the DRIS indices increases. Sumner (10) stated that the potential yield attainable increases as the sum of the DRIS indices decreases. Low yields and low sums suggests that other factors were limiting yields.

Table 6. Correlation coefficients, and probability levels relating cane yields to the nutrient content of leaves and sheath moisture.

87

Figure 1. Relationship between leaf K concentration and sheath moisture level of 3-month-old sugarcane.

In the surveyed sugarcane the absolute values of the indices were small, less than 100 throughout the season, and were not correlated with cane yields. The average absolute values of the DRIS indices for fields with high or low yield records were 10.8 and 17.8. No significant relationships were found between cane yields and the DRIS indices for N, K, P or sheath moisture of 3 month (Figure 2) or older cane.

Figure 2. Relationship between relative yield (% of maximum) and N, P, K or sheath moisture indices of 3-month-old sugarcane.

Apart from the effect of moisture stress on the plant's nutrient composition, the moisture status of the plant is a critical factor affecting growth and yield. A modified DRIS approach was used to evaluate the nutrient-moisture relationship. DRIS indices based on ratios involving N, P, and sheath moisture (M) suggested that the moisture status was the factor limiting crop growth in 82% of the fields of 3-month-old cane surveyed (Table 7). Based on a normal sheath moisture level of 83%, low tissue moisture was a limiting factor in 53.6% of the surveyed fields.

Juice quality expressed as pol and purity was not significantly influenced by the concentration of N, P, K, Ca or Mg in the leaves of 6-month-old cane. Other studies (12, 15) have reported that juice quality was significantly affected by the moisture status of the plant at harvest and that excessive N availability during the

88

ripening period decreases juice quality. However, the low leaf N and sheath moisture levels of the 6-month-old cane suggested that N and moisture were not major factors affecting the juice quality of the surveyed cane.

Table 7. Classification of nutrient requirements by DRIS based on ratios involving N, P, K, Ca, Mg and sheath moisture (M).

CONCLUSIONS

Leaf N concentration of 3-month-old cane suggested that 38% of the surveyed fields needed a supplemental application of N fertilizer. Neither P nor K appeared to be serious problems. Leaf P and K concentrations were normal or higher in 89 and 92%, respectively, of the fields. The low P and K levels probably reflected low availability of N or water. Low absolute DRIS index values indicated that nutrient imbalance was a minor factor affecting cane growth.

REFERENCES

1. Beaufils, E. R. 1973. Diagnosis and recommendations integrated system (DRIS). A general scheme for experimentation and calibration based on principles developed from research in plant nutrition. Soil Sci. Bul. No. 1., Univ. of Natal, S. Africa.

2. Beaufils, E. R., and M. S. Sumner. 1976. Applications of the DRIS approach for calibrating soil and plant factors in their effects on yield of sugarcane. Proc. S. Africa Sugar Tech. Assoc. 50:118-124.

3. Bolin, O. W., and O. E. Stramberg. 1944. Rapid digestion method for determination of phosphorus. Ind. Eng. Chem. Am. Ed. 16:345-346.

4. Bremmer, J. M. 1965. Total nitrogen. pp. 1149-1176. In C. A. Black (ed) Methods of soil analysis, Part 2. Am. Soc. Agm, Madison, Wisconsin.

5. Clements, H. F. 1980. Sugarcane crop logging and crop control: principles and practices. The University Press of Hawaii, Honolulu.

89

6. Elwali, M. D. and G. J. Gascho. 1983. Foliar critical nutrient levels and "DRIS" norms as guides for sugarcane fertilization. Proc. Inter. Amer. Sugar Cane Seminars, Soil Fertility and Management. pp. 20-35.

7. Escano, C. R., C. A. Jones, and G. Vehara. 1981. Nutrient diagnosis in corn grown on Hydric Dystrandepts: II. Comparison of two systems of tissue diagnosis. Soil Sci. Soc. Amer. J. 45:1140-1144.

8. Hipp, B. W. 1969. Potassium fixation and supply by soils and mixed clay minerals. Tex. Agric. Exp. Sta. B-1090.

9. Richardson, A. J. 1984. El Chicon Volcanic ash effects on atmospheric haze measured by NOAAI AUHRR data. Remote Sensing of Environment. 16:157-164.

10. Sumner, M. S. 1977. Application of Beaufil's diagnostic indices to maze data published in the literature irrespective of age and conditions. Plant Soil 46:359-369.

11. Thomas, J. R., F. G. Salinas, and G. F. Oerther, Jr. 1974. Ratoon chlorosis of sugarcane. J. Rio Grande Valley Hortic. Soc. 28:169-174.

12. Thomas, J. R., and G. F. Oerther, Jr. 1976. Growth, production, and leaf N content of sugarcane in Texas. Proc. ASSCT. 5:28-36.

13. Thomas, J. R., F. G. Salinas, and L. N. Namken. 1978. Growth and yield of sugarcane as affected by row spacing and irrigation regime. Proc. ASSCT. 7(NS):129-135.

14. Thomas, J. R. 1983. Calibration of leaf N percentages for predicting the N fertilizer needs of sugarcane. Proc. Inter. Amer. Sugar Cane Seminar. Soil Fertility and Management. pp. 95-103.

15. Thomas, J. R., A. W. Scott, Jr., and R. P. Wiedenfeld. 1985. Fertilizer requirements of sugarcane in Texas. Jour. ASSCT 4:62-72.

16. Thomas, J. R. and N. Rozeff. 1988. Investigation of potassium needs of sugarcane in Texas. Jour. ASSCT 8:38-43.

90

GROWTH RESPONSE OF SIX SUGARCANE CULTIVARS TO THE HERBICIDES ASULAM, DALAPON AND MSMA

R. W. Millhollon and H. P. Fanguy Sugarcane Research Unit, ARS, USDA

Houma, Louisiana 70361

ABSTRACT

The commercial sugarcane cultivars CP 65-357, CP 70-321, CP 72-356, CP 72-370, CP 73-351, and CP 74-383, interspecific hybrids of the genus Saccharum, were evaluated in two Louisiana field experiments for their tolerance to foliar applications of the postemergence herbicides asulam at 3.7 and 6.7 kg/ha, dalapon at 5.0 kg/ha and MSMA at 4.5 kg/ha. Herbicides were applied in late April to new spring growth of sugarcane as would be done for control of johnsongrass, but cultivars were maintained weed-free. Cultivar response to the herbicides was determined from early-season shoot height, mature stalk production, and cane yield. Cultivars were not injured by asulam at the standard rate of 3.7 kg/ha. Asulam at 6.7 kg/ha generally caused more initial reduction in plant height and leaf chlorosis, but only CP 72-370 had reduced yield and this in only one experiment Dalapon reduced early-season plant height for all cultivars, but its effect on number of mature stalks and cane yield at harvest varied with cultivar: CP 72-370 and CP 65-357 were tolerant; CP 73-351 and CP 74-383 had some adverse reaction in one or both experiments; and CP 70-321 and CP 72-356 had consistent reductions in both experiments. MSMA caused temporary leaf desiccation, and generally reduced the early-season growth of all cultivars, but it did not affect mature stalk production for any cultivar. CP 74-383 and CP 72-370 exhibited greater tolerance to MSMA than the other four cultivars which had significant yield reductions in one or both experiments. These herbicide tolerance characterizations for the six cultivars provide information required for the development of efficient weed-management systems in sugarcane.

INTRODUCTION

In Louisiana, johnsongrass [Sorghum halepense (L.) Pers.], itchgrass [Rottboellia cochinchinensis (Lour. Clayton] and a complex of other grass and broadleaved weeds in sugarcane are routinely controlled with herbicides. Johnsongrass from rhizomes and itchgrass can be particularly troublesome in the ratoon crops. Overhead applications of asulam [methyl (4-aminophenyl) sulfonyl carbamate] can be used effectively for control, usually without significant injury to sugarcane (5, 7). Although not used as extensively as asulam, dalapon (2,2-dichloropropanoic acid) can be used as an overhead treatment for control of johnsongrass, itchgrass, bermudagrass [Cynodon dactylon (L.) Pers.] and annual grasses (4, 5, 7). MSMA (monosodium salt of methylarsonic acid) is not currently registered for use in sugarcane in the U. S., but it effectively controls johnsongrass and itchgrass (4, 7) and is used widely for sugarcane weed control in other countries. Overhead treatments with dalapon and MSMA can injure sugarcane, but injury is minimized by applying these herbicides in early spring when temperatures are cool and sugarcane growth is slow (4, 5).

In Louisiana three or more cultivars of sugarcane are grown on individual farms to take advantage of the best qualities of each, although any one field is planted to only one cultivar. Cultivars are replaced as higher-yielding, more disease-resistant ones are developed (1). Cultivars are known to respond differentially to herbicides: CP 44-101 and CP 52-68 were more tolerant than NCo 310 and L 60-25 to dalapon (2, 8); NCo 310 was more tolerant than CP 44-101 and CP 52-68 to diuron [N'-(3,4-dichlorophenyl)-N,N-dimethylurea] (8); CP 74-383, CP 73-351, and CP 72-356 were more tolerant than CP 48-103, CP 65-357, and CP 72-370 to terbacil [5-chloro-3-(l,l-dimethylethyl)-6-methyl-2,4(lH,3H)-pyrimidinedione] (9); and CP 72-356, CP 73-351, CP 74-383, and CP 70-321 were more tolerant than CP 65-357 and CP 72-370 to hexazinone [3-cyclohexyl-6-(dimethylamino)-l-methyl-l,3,5-triazine-2,4(lH,3H)-dione] (6, 9). The degree of tolerance to certain herbicides becomes an important characteristic of the cultivar.

The purpose of this study was to characterize the response of six currently important commercial cultivars to asulam, dalapon and MSMA.

91

MATERIALS AND METHODS

Field experiments were conducted in 1983 and repeated in 1986. The sugarcane used for treatments was the spring growth of cane first harvested the previous autumn (first-ratoon crop). It had been planted in plots (5.2 m x 5.5 m) about 20 months before treatment on Mhoon silt loam soil at the U. S. Sugarcane Field Laboratory's Ardoyne Farm near Houma, Louisiana. The experiments were herbicide treatment by cultivar factorials arranged in a split-plot design with four (1983) or three (1986) replications. Asulam at 3.7 or 6.7 kg/ha, dalapon at 5.0 kg/ha, MSMA at 4.5 kg/ha, and untreated control were whole plots, and commercial cultivars - CP 65-357, CP 70-321, CP 72-356, CP 72-370, CP 73-351, and CP 74-383 - were subplots arranged in randomized complete blocks. Herbicide rates are expressed as the amount of active ingredient (asulam and MSMA) or acid equivalent (dalapon) applied broadcast to a hectare of sugarcane.

Plots were maintained weed free by treating with atrazine [6-chloro-N-ethyl-N'-(l-methylethyl)-l,3,5-triazine-2,4-diamine] at 2.2 kg/ha following planting and each spring thereafter and by supplementary hand weeding as required. Experiments have shown that such atrazine treatments are not phytotoxic to sugarcane (R. W. Millhollon, unpublished data). Herbicide treatments were applied on April 26, 2983 (Exp. 1) or May 6, 1986 (Exp. 2); the cane ranged from 51 to 76 cm in overall height at the time of treatment. Commercial formulations of the sodium salt of asulam, MSMA, or the sodium salt of dalapon were applied over-the-top of sugarcane foliage in water sprays of 380 1/ha containing a 0.25% (v/v) commercial nonionic surfactant. Herbicides were applied on a 90-cm band over rows 80 cm wide so that all sugarcane leaves were wetted. The mean daily maximum and minimum temperatures for the period two weeks before to two weeks after herbicide treatment were 24.5 C and 16.0 C for Experiment 1 and 29.0 C and 18.8 C for Experiment 2. Rainfall of about 5.0 cm was recorded for each experiment within this same period but no rainfall occurred within 24 hours of treatment.

The effect of treatments on growth of cane was determined from shoot heights taken about 30 days after treatment. In Experiment 1, six randomly selected shoots per plot were measured from the ground to the top of the leaf canopy (overall height), whereas in Experiment 2, ten randomly selected shoots/plot were measured from the ground to the top visible dewlap. The number of harvestable stalks/plot (those more than about 1.2 m tall) was counted in September; the weight of cane/plot was determined in early November when the cane was cut mechanically and burned. A 15-stalk sample of the cut cane was randomly selected from each plot and crushed in a sugarcane sample mill to extract the juice; the juice was analyzed for sucrose content using standard methods (3).

The data was analyzed as a factorial experiment in a split-plot design using standard analysis of variance (ANOVA) procedures to determine significant differences (P = 0.05) for whole plots (herbicides), sub plots (cultivars) and the interactions. When the interaction was significant, further analyses were made using single degrees of freedom to test for differences (P = 0.05) among cultivars after subtracting the value of the control. As an example, t/ha for the CP 65-357 control - t/ha for dalapon-treated CP 65-357 was compared with t/ha for the CP 74-383 control - t/ha for dalapon-treated CP 74-383. These data are presented as "% of control" in the tables for clarity. Interactions were interpreted from the "% of control" data, which placed all cultivars on an equal basis for comparison, and from the significant differences determined for treatment means as compared to the control.

RESULTS AND DISCUSSION

Because of variation between experiments, data for shoot height 30 days after treatment (Table 1), number of harvestable stalks produced (Table 2), and yield as weight of cane (Table 3) are presented separately for Experiments 1 and 2. Sugar content of stalks is not presented because of the lack of significant differences in the ANOVA for herbicide treatments or for interactions. Herbicide treatment by cultivar interactions were significant for stalk number and yield of cane in each experiment, but no significant interaction was found for early-season shoot height.

All cultivars showed good tolerance to asulam at 3.7 kg/ha, the standard rate, as measured by growth and yield parameters in both Experiment 1 and 2 (Tables 1, 2, and 3). The 6.7 kg/ha rate of asulam generally was more phytotoxic than the lower rate, reducing shoot height below the control as an average of all cultivars (Table 1). In addition to shoot height reduction, leaf chlorosis was also observed for all cultivars in

92

Experiment 2. CP 72-370 was observed to be much more chlorotic than other cultivars in Experiment 2, and this injury, while not greatly affecting stalk number (Table 2), was associated with a significant reduction in yield of cane as compared to the control (Table 3). Yields for other cultivars were not significantly different from the control.

Table 1. Shoot height of sugarcane cultivars 30 days after application as affected by foliar herbicide treatments.

1 Analysis of variance for each experiment showed no significant herbicide treatment by cultivar interaction; thus, significant differences are only indicated for herbicide treatment means calculated as an average of all cultivars.

2 Means followed by the same letter are not significantly different (P = 0.05) as determined by Duncan's multiple range test. 3 Height was measured from ground to top of leaf canopy. 4 Height was measured from ground to uppermost dewlap.

The difference in reaction of cultivars to asulam in the two experiments show that asulam can be quite variable in causing injury to sugarcane. Extensive observations by the authors have shown that asulam is most likely to injure sugarcane in Louisiana when it is applied from the middle of May through July, a period of both relatively high temperatures and rapid sugarcane growth. In addition to the high temperature, other stresses that appear to favor injury are drought and severe weed competition. The high rate of asulam in this study was used in an attempt to induce injury so that differences in tolerance between cultivars could be detected. CP 72-370 was injured more than other cultivars by the high asulam rate, and this response agrees with observations in commercial fields.

Dalapon reduced shoot height of all cultivars and the magnitude of the reduction appeared to be similar in both Experiment 1 and 2 (Table 1). However, dalapon affected stalk production and yield of CP 73-351 more in Experiment 2 than in Experiment 1 (Tables 2 and 3). Although the "% of control" figures generally were not significantly different among cultivars for stalk production and cane yield, CP 70-321 and CP 72-356 produced fewer stalks and lower cane yield than the controls in both experiments, indicating that they were relatively sensitive to the dalapon treatment. Similar reductions were found for CP 74-383 except for cane yield in Experiment 2. Yields for CP 72-370 and CP 65-357 were not reduced following dalapon treatment.

93

Table 2. Mature stalk production of sugarcane cultivars as affected by foliar herbicide treatments.

1 Lower case letters are for comparison of herbicide treatment means (no/ha) within a cultivar; upper case letters are for comparison of cultivars (no/ha as % of the control) within a herbicide treatment. Means followed by the same letter are not significantly different (P = 0.05) as determined by the Duncan's multiple range test for means within a cultivar or by an LSD analysis for cultivar comparisons within a herbicide treatment.

A comparison of the "% of control" data for the effect of MSMA on growth of shoots for each cultivar indicates that MSMA was generally more phytotoxic in Experiment 2 than in Experiment 1 (Table 1). The injury caused by MSMA did not greatly affect stalk production (Table 2), but did affect yield of cane for CP 65-357, CP 70-321, CP 72-356, and CP 73-351 in one experiment as shown by the comparison of actual yields with the controls (Table 3). CP 74-383 and CP 72-370 were not affected by MSMA treatment in either experiment.

94

Table 3. Yield of sugarcane cultivars as affected by foliar herbicide treatments.

1 Lower case letters are for comparison of herbicide treatment means (t/ha) within a cultivar; upper case letters are for comparison of cultivars (t/ha as % of the control) within a herbicide treatment. Means followed by the same letter are not significantly different (P = 0.05) as determined by the Duncan's multiple range test for means within a cultivar or by an LSD analysis for cultivar comparisons within a herbicide treatment.

This study showed that dalapon and MSMA are more phytotoxic to sugarcane than the standard asulam treatment. It also showed that the cultivars in this study varied in their tolerance to asulam, dalapon and MSMA. Such knowledge is of value to sugarcane producers because the objective of an efficient herbicide weed control program is to selectively control weeds so that a sugarcane cultivar can achieve near to its maximum yield potential.

REFERENCES

1. Fanguy, H. P. and D. B. Fontenot. 1987. Louisiana's 1986 sugarcane variety census. Sugar y Azucar 82:31-33.

2. Matherne, R. J. and R. W. Millhollon. 1973. Tolerance of two sugarcane cultivars to terbacil, fenac and dalapon. Weed Sci. 21:139-141.

3. Meade, G. P. and J. C. P. Chen. 1977. Cane Sugar Handbook (10th ed). Wiley-Interscience Publications. John Wiley and Sons. New York, NY. 947 p.

4. Millhollon, R. W. 1970. MSMA for johnsongrass control in sugarcane. Weed Sci. 18:333-336.

5. Millhollon, R. W. 1976. Asulam for johnsongrass control in sugarcane. Weed Sci. 24:496-499.

6. Millhollon, R. W. 1986. Factors affecting tolerance of sugarcane (Saccharum officinarum) to hexazinone. Jour. ASSCT 6:5-10.

95

7. Millhollon, R. W. 1986. Control of itchgrass [Rottboellia cochinchinensis (Lour.) Clayton] in sugarcane with post-emergence herbicide treatments. Proc. ISSCT 19:80-91.

8. Millhollon, R. W. and R. J. Matherne. 1968. Tolerance of sugarcane varieties to herbicides. Weed Sci. 16:300-303.

9. Richard, E. P., Jr. 1989. Response of sugarcane (Saccharum sp.) cultivars to preemergence herbicides. Weed Technol. 3:358-363.

96

EFFICIENCY OF IN VITRO PROPAGATION OF SUGARCANE PLANTS BY DIRECT REGENERATION FROM LEAF TISSUE AND BY SHOOT-TIP CULTURE

Michael P. Grisham and Druis Bourg Sugarcane Research Unit, ARS, USDA

Houma, Louisiana 70361

ABSTRACT

The efficiency of in vitro propagation of sugarcane cultivars CP 65-357 and CP 70-321 by direct regeneration from leaf tissue and shoot-tip culture was studied. Plant production by direct regeneration from leaf roll sections was low (4 to 18 plants/leaf roll). In shoot-tip culture, 85 and 90% of the initial shoot-tips of CP 65-357 and CP 70-321 were advanced to the shoot-proliferation stage, respectively. The estimated production by shoot-tip culture for six transfers in the shoot-proliferation medium was over 27,500 plants for CP 65-357 and over 7,000 plants for CP 70-321. The best liquid rooting medium formulation tested for cultivar CP 65-357 contained 1/2 concentration of Murashige and Skoog salts and 60 g/1 sucrose. Three regimes of daylength and temperature were compared for ability to maintain in vitro cultures without transfer.

INTRODUCTION

Micropropagation of sugarcane offers the opportunity to produce large numbers of disease-free plants for commercial and research applications provided that the donor plants are disease-free (1). Two in vitro methods suggested for the rapid propagation of sugarcane are direct regeneration of plants from leaf tissue (3) and plant propagation from shoot-tip cultures (2). The purpose of this study was to compare the efficiency of the direct regeneration method as described (3) with a modified shoot-tip culture procedure for the rapid propagation of two leading Louisiana commercial cultivars of sugarcane. Modifications of the rooting medium of the shoot-tip culture and conditions of daylength and temperature for maintenance of in vitro cultures are reported.

MATERIALS AND METHODS

Two in vitro methods were used to propagate plants of sugarcane cultivars CP 65-357 and CP 70-321. Initial success rate to establish in vitro culture, time required to produce plants capable of surviving in the field, and the propagation rate of plants produced per original culture were determined.

In vitro cultures were maintained at 26 C under light which consisted of indirect sunlight supplemented with fluorescent lights (Westinghouse Agro-Lite1, 20 watt, about 20 cm above the vessels). Plants were maintained in the greenhouse for two weeks to three months before transplanting to the field.

Direct regeneration - Apical portions of healthy stalks of sugarcane cultivars CP 65-357 and CP 70-321 were stripped, surface sterilized, and the immature leaf roll just above the apical meristem cut transversely into six to eight, 3 mm-thick sections. These were placed on agar medium containing Murashige and Skoog (MS) salts (6), 20 g/1 sucrose, 2 mg/1 kinetin, and 5 mg/1 naphthaleneacetic acid (3). As shoots arose from the cut surface of the leaf roll tissue, they were separated and cultured on agar medium containing MS salts, 20 g/1 sucrose, and 2 mg/1 indolebutyric acid (IBA) until roots were formed (3). Plants were placed in Model 200 Todd planter flats (5.2 x 5.2 x 7.6 cm cavities) (Speedling, Inc., Sun City, FL)1 filled with vermiculite and kept in the laboratory for 24-48 hours before being moved to the greenhouse.

1Mention of a trademark or proprietary product does not constitute a guarantee or warranty of the product by the U. S. Department of Agriculture, and does not imply its approval to the exclusion of other products that may also be suitable.

97

Shoot-tip culture - The shoot tip (approximately 4 mm long x 4 mm in dia) was asceptically excised from the apical portion of healthy stalks by completely removing the leaf roll tissue. The shoot tip was placed in 20 ml of broth containing MS salts, 14 mg/1 of a commercial, 15-30-15, fertilizer (Miracle-Gro, Stern's Miracle-Gro Products, Inc., Port Washington, N. Y.)1, 20 g/1 sucrose, 0.1 mg/1 gibberellic acid, and 0.01 mg/1 IBA to stimulate shoot elongation (2). Shoot tips were unsupported in the Magenta GA-7 Vessels (7.6 x 7.6 x 10.2 cm) (Magenta Corporation, Chicago, IL 60641)1 and were agitated on an orbital shaker at 40 revolutions per minute.

After shoot tips reached 10-20 mm in length, they were transferred to 20 ml of broth containing MS salts, 70 mg/1 of Miracle-Gro, 20 g/1 sucrose, 0.1 mg/1 kinetin, and 0.2 mg/1 benzylaminopurine for shoot proliferation (2, as amended by S. Kresovich, personal communication). At intervals of 8 to 24 days, proliferating shoots were separated into clusters with a range of 16 to 52 shoots each, and each cluster was placed in fresh shoot-proliferation medium. Multiplication was continued until the desired number of shoots was obtained. Records were kept on four clusters of two original shoot tips per transfer per cultiver.

Following shoot proliferation, clusters were placed in 20 ml of broth containing MS salts, 34 mg/1 Miracle-Gro, 20 g/1 sucrose, and 2 mg/1 IBA to induce roots (3). When roots developed, plants were separated, potted individually in planter flats filled with vermiculite, and placed in the greenhouse. The propagation rate was calculated.

Rooting medium - Four rooting media in which MS salts, sucrose, and IBA concentrations differed were compared using clusters of shoots of CP 65-357. Each cluster was placed in 20 ml of medium after recording the number, size, and appearance of the shoots. Each cluster was transferred to fresh medium after two weeks and shoot number, length, and vigor and root number and length were recorded four weeks after initiation of experiment. Each treatment was replicated three times and arranged in a completely randomized design. In a second rooting study, the best medium from the first study was compared to three other media. Each treatment was replicated ten times and arranged in a completely randomized design.

Effect of davlength and temperature on maintenance of in vitro cultures - Shoots of CP 65-357 and CP 70-321 in clusters of approximately 20 shoots each were added to culture vessels containing 20 ml of the shoot-proliferation medium. Twenty vessels of each cultivar were placed in separate incubators set at 1) a diurnal cycle of 8 h: 16 h, day (21 C); night (12 C); 2) 8 h: 16 h, day: night (constant 20 C); and 3) 16 h: 8 h, day, night (constant 20 C). Each set of conditions was repeated in a different incubator with new cultures of each cultivar. A subjective rating of shoot condition was made at 2, 4, 6, 8 and 16 weeks.

RESULTS AND DISCUSSION

Plants produced by both micropropagation methods formed sufficient roots after four weeks in the rooting medium to be transferred to planter flats containing vermiculite and to be placed in the greenhouse. More than 90% of the plants survived transfer to the greenhouse and, again, to the field.

Direct regeneration - The number of plants produced by the direct regeneration method is considered very low (Table 1). Plantlets were recovered from about 25% of the leaf rolls cultured. Most of the leaf-roll sections that failed to produce plantlets deteriorated within 60 days of culture. Microbial contamination did not appear to be a factor in culture establishment. The time required to initiate plantlets directly from the leaf-roll tissue ranged from 75 to 261 days with most plantlets formed within the first 125 days.

Shoot-tip culture - Of the 13 shoot tips from CP 65-357 and 20 from CP 70-321 used to initiate shoot-tip cultures, 85 and 90% were successfully advanced to the shoot-proliferation stage, respectively. Microbial contamination caused the loss of a few cultures. Shoot proliferation of two shoot tips of each cultivar is given in Table 2. In shoot-tip culture, elongation took 21 days. Following transfer of regenerated plants to the shoot-proliferation medium, the first division of shoot clusters of CP 65-357 was made at 21 days (Table 2); while clusters of CP 70-321 shoots required a transfer to fresh proliferation medium at 21 days and the first division of clusters 20 days later (Table 2).

98

Table 1. Production of sugarcane plants of cultivars CP 65-357 and CP 70-321 by direct regeneration from leaf roll sections.

1 As shoots were produced, they were separated from the leaf roll tissue and transferred to rooting media; other pieces of leaf roll tissue were transferred to fresh shoot initiation media. Plant initiation: initiated plants had sufficient roots to survive transfer to greenhouse.

Table 2. Shoot-tip culture multiplication of sugarcane shoots of cultivars CP 65-357 and CP 70-321 in shoot-proliferation medium.

1 Shoot multiplication was recorded for two shoot tips of each cultivar. Four clusters of shoots per tip per cultivar were advanced in each transfer.

99

Once roots formed on shoot-tip clusters, individual plants could be separated. Some shoot proliferation continued in the rooting medium. Rooted plants from a cluster which could not be readily separated were planted together. After approximately two weeks growth, they could be separated and planted individually.

The plant production time and the number of plantlets produced per leaf roll by the direct regeneration procedure (Table 1) were similar to those reported by Irvine and Benda (3). Cultures of CP 70-321 were less responsive to direct regeneration than cultures of CP 65-357. The rate of multiplication of CP 65-357 in shoot-tip procedure (Table 2) was similar to that reported by Hendre et al (2) for cultivar Co 740; however, the rate was less than reported by Lee (5) for cultivar RB 735275. A lower number of plants was produced in cultures of CP 70-321 after a comparable amount of time because an additional transfer was needed before shoot proliferation was sufficient to make the first division of the shoot clusters and because the average time between transfer and division of clusters was greater. Kresovich et al (4) reported similarly that CP 70-321 was less responsive than NCo 310 to callus culture. Using the multiplication rates in Table 2, the estimated production by shoot-tip culture from a single shoot tip after six transfers in shoot-proliferation medium was approximately 27,500 plants for CP 65-357 and 7,000 plants for CP 70-321 and would take approximately four months for CP 65-357 and five months for CP 70-321 from the time shoot-tips were collected till rooted plants were placed in the greenhouse.

Shoot-tip culture offers the ability to produce a large number of plants from a single shoot tip because the shoot-proliferation stage can be repeated many times. There is the potential for abnormal plant or tissue formation after many cycles of multiplication. Although no abnormal plants were observed after six multiplication cycles in this study, Lee (5) observed a green mass at the base of formed shoots after seven cycles. In the direct regeneration method, as previously described (3) and used here, plantlets are taken directly from the leaf roll and placed in rooting medium. Further plant production requires the initiation of new leaf-tissue cultures. If large numbers of plants were desired from direct regeneration, the transfer of newly formed plantlets on the leaf roll to a shoot-proliferation medium could be attempted.

The commercial fertilizer added to shoot-proliferation medium of the shoot-tip procedure eliminated any evidence of stress on the shoots during two-week or longer transfers. Without the supplemental nutrients, at 7-10 days after transfer, leaf tips began to brown and older leaves became chlorotic; therefore, shoot clusters were transferred to fresh medium before they were ready to be subdivided. Quantity of the shoot-proliferation medium did not appear to be the limiting factor since increasing the volume of the original medium did not eliminate the need for the additional transfer (unpublished).

Rooting medium - In the first study (Table 3), increasing the sucrose content, reducing the concentration of the MS salts, and eliminating IBA resulted in the best root development. The second study demonstrated that shoots in root-inducing media containing 60 and 90 g/1 sucrose produced approximately the same number of roots; however, the roots produced at the higher concentration were slightly discolored (Table 4). Indolebutyric acid at 2 mg/1 reduced shoot and root vigor, while IBA at .2 mg/1 appeared to have little or no effect on rooting when combined with the increased sucrose concentration (Table 4).

Formulation H (Table 4) which contained 60 g sucrose and no IBA was the best of these formulations for the micropropagation of CP 65-357, and has been used for micropropagation of CP 70-321, CP 74-383, and CP 79-318. Roots were obtained in two weeks using formulation H compared to four weeks with formulation A in the above experiments.

Effect of davlength and temperature on maintenance of in vitro cultures - Shoot tip elongation and initiation of proliferation took from five to eight weeks; therefore, if cultures were available at the shoot proliferation stage, considerable time would be saved. With efficient maintenance procedures, the preservation of a collection of cultivars in vitro would be facilitated.

100

Table 3. Effect of rooting-medium formulations on shoot and root development of shoot clusters of CP 65-357 (first study).

Ave. Ave. Ave. Ave. shoot shoot root root

no./ length Shoot no./ length MS1 salts Sucrose IBA1 cluster (mm) vigor2 cluster (mm)

Formulation concentration (g/1) (mg/1) start 4 wk start 4 wk 4 wk 4 wk 4 wk

A IX 20 2 8 49 18 43 10 17 15 B 1/2X 90 0 8 29 19 42 10 51 23 C IX 90 2 9 16 17 16 4 20 8 D 1/2X 90 2 8 23 18 23 5 50 13

1 MS = Murashige and Skoog (6); IBA = indolbutyric acid. 2 Scale of 1-10, with 1 = very poor shoot condition, survival doubtful and 10 = excellent shoot vigor, no evidence

of stress. All shoots initially rated 10.

Table 4. Effect of rooting-medium formulations on shoot and root development of shoot clusters of CP 75-357 3 wk after transfer to medium (second study).

Formulation

E F G H B

MS1 salts concentration

IX 1/2X 1/2X 1/2X 1/2X

Sucrose (g/1)

0 60 60 60 90

IBA1

(mg/1)

2.0 0.2 2.0 0.0 0.0

Ave. shoot vigor2

8 7 6

10 9

Ave. shoot no./

cluster

23 24 18 26 23

Ave. root no./ cluster

0 >50

3 >50 >50

Root appearance Color

white brown white slightly brown

Vigor

vigorous growth poor growth vigorous growth vigorous growth

1 MS = Murashige and Skoog (6); IBA = indolbutyric acid.

2 Scale of 1-10, with 1 = very poor shoot condition, survival doubtful and 10 = excellent shoot vigor, no evidence of stress. All shoots initially rated 10.

In the 8 h: 16 h, day (21 C): night (12 C) diurnal cycle, some growth and tillering continued although at a much reduced rate. Shoot counts doubled in both cultivars after eight weeks. Although cultures were extensively yellowed, all cultures remained viable after 16 weeks without transferring to fresh medium (Table 5). Most cultures would probably have survived longer. Under constant temperature of 20 C, length of the light period became important (Table 5). A longer light period was needed to maintain green shoots longer than ten days.

Shoot-tip culture offered a clear advantage as a method for large-scale propagation of sugarcane. In the subsequent propagation of three cultivars, it was determined that the use of the shoot-elongation medium was unnecessary and shoot tips could be placed directly into the shoot-proliferation medium. There also appears to be no advantage in the direct-regeneration method of propagation for freeing plants from sugarcane mosaic virus (3).

101

Table 5. Maintenance of clusters of CP 65-357 and CP 70-321 shoots in shoot-proliferation medium as affected by photo- and thermo-period.

Treatment Light Dark Condition of shoots1

Hours Temp(C) Hours Temp (C) Cultivar 2 wk 4 wk 6 wk 8 wk 16 wk

8 21 16 12 CP 65-357 G G G MY EY,T CP 70-321 G G MY MY EY,T

12 20 12 20 CP 65-357 G G MY MY,T CP 70-321 G G MY MY,T

8 20 16 20 CP 65-357 MY MY MY EY,T CP 70-321 MY MY EY EY,T

1 Condition of shoots: G = green, healthy appearing shoots, no symptom of stress. MY = moderate yellowing, yellowing beginning with older leaves. EY = extensive yellowing, only youngest leaves green, obvious symptoms of stress. T = transferred and all cultures were viable.

Plants produced by both methods are currently being evaluated in the field for agronomic characters. A large number of plants produced by shoot-tip culture are currently being used in a study of sugarcane mosaic spread because they are a population of plants from a common, disease-free source and could be space planted to maintain individual stool integrity.

ACKNOWLEDGEMENTS

The authors are grateful to Messrs. Southern Bibbins and Calvin Tolbert for their careful maintenance of plants in the greenhouse and field.

REFERENCES

1. Carlson, P. S. 1983. Uses of cell and tissue culture for sugar cane improvement. Sugar y Azucar 78(6):33.

2. Hendre, R. R. R. S. Iyer, M. Kotwal, S. S. Khuspe, and A. F. Mascarenhas. 1983. Rapid multiplication of sugar cane by tissue culture. Sugar Cane (l):5-8.

3. Irvine, J. E., and G. T. A. Benda. 1985. Sugarcane mosaic virus in plantlets regenerated from diseased leaf tissue. Plant Cell, Tissue and Organ Culture 5:101-106.

4. Kresovich, S., R. E. McGee, and S. J. Wadsworth. 1985. Comparisons of the in vitro responsiveness of callus cultures derived from immature inflorescence and leaf base tissues of two interspecific hybrids of sugar cane (Saccharunt spp.). Sugar Cane (6):6, 9-10.

5. Lee, T. S. G. 1987. Microprogation of sugarcane (Saccharunt spp.). Plant Cell, Tissue and Organ Culture 10:47-55.

6. Murashige, T., and F. Skoog. 1962. A revised medium for rapid growth and bio assays with tobacco tissue cultures. Physiol. Plant. 15:473-497.

102

ABSTRACTS - AGRICULTURE

EFFECTS OF BY-PRODUCT GYPSUM ON YIELD AND NUTRIENT CONTENT OF SUGAR

CANE AND SOIL PROPERTIES

J. A. Breithaupt, Allen Arceneaux, and Ray Ricaud Agronomy Department

Louisiana Agricultural Experiment Station Louisiana State University Agricultural Center

Baton Rouge, Louisiana

An experiment was conducted to determine the effects of fluorogypsum and phosphogypsum (CaS04 • 2HzO) on the yield and nutrient content of sugar cane and chemical properties of the soil.

By-product gypsum was applied to a Sharkey clay soil (Vertic Haplaquept, very fine, montmorillonitic, thermic, non-acid) at rates of zero, one, two, five, and ten tons per acre of fluorogypsum and five tons per acre of phosphogypsum. Samples were taken from the Ap, AC, and C soil horizons in the plant cane and first stubble crop years.

Cane and sugar yields increased with each treatment and the ten tons per acre fluorogypsum significantly increased yield over the check plot in the plant cane. The five and ten tons per acre fluorogypsum and five tons per acre phosphogypsum significantly increased yield over the zero and one ton per acre fluorogypsum treatments in the first stubble crop year.

There were no significant differences in nutrient content of plant tissue in the plant cane year. However, S content in plant tissue with the ten tons per acre treatment was significantly higher than the zero and one tons per acre fluorogypsum in the first stubble year.

Extractable soil sulfur was significantly higher with the five and ten tons per acre fluorogypsum than with the check in both crop years. Extractable soil Ca increased significantly with the ten tons per acre fluorogypsum treatment over the other treatments. Extractable soil Mg was significantly lower the five tons per acre than the one tone per acre of fluorogypsum in the plant cane. Significant differences were obtained among treatments in some of the trace elements.

CORRELATION OF CROP AGE WITH POPULATIONS OF SOIL INSECT PEST IN FLORIDA SUGAR CANE

Ron Cherry Everglades Research and Education Center

Belle Glade, Florida

Correlation between crop age and populations of soil insect pests was measured in 18 commercial sugar cane fields in Florida. Melanotus communis (Gyllenhal) was the most abundant and largest wireworm species found in these fields. Wireworm populations were not significantly correlated with crop age (years) as indicated by a low correlation coefficient of -0.18. Ligyrus subtropicus (Blatchley) was the most abundant and largest grub species found in these fields. In contrast to wireworms, grub populations were significantly correlated with crop age (years) with a correlation coefficient of +0.74. Data presented in this study indicate the importance of old sugarcane fields in harboring grub populations and these data also suggest reduced ratooning of Florida sugar cane as a possible means of grub control.

103

RATOON STUNTING DISEASE LOSSES IN FOUR COMMERCIAL SUGAR CANE CLONES IN FLORIDA

J. L. Dean, M. J. Davis, and N. A. Harrison University of Florida

Fort Lauderdale, Florida

Yield loss due to ratoon stunting disease (RSD) of sugar cane was measured at four locations, two on sand and two on muck, in plant cane and first ratoon. At each location there were eight replications in a randomized-complete-block, split-plot design with clones as main plots and disease states (healthy or RSD infected) as sub-plots. Each sub-plot was four rows x 5.3 m surrounded by 4.6 m of clear space on four sides.

All seed cane for the trials came from a nursery at Canal Point, Florida. The nursery had been established by treating all seed cane in hot water at 51°C for two hours, then inoculating seed for half of the plots with Clavibacter xyli subsp. xyli from culture. The infection status of seedcane from plots in the increase nursery was established by examination of extracted sap for the presence or absence of CX. subsp. xyli by light microscopy.

The variance of three parameter (sugar per tonne of cane, tonnes of cane per ha, and tonnes of sugar per ha) was analyzed for each location separately each year and for all locations and years combined. In the separate analyses,one or more clones showed a significant loss (P=0.5) of sugar per ha due to RSD in each trial, but with little consistency with respect to clones from trial to trial. In the combined analysis, all four clones showed a significant loss in both tonnes of cane and sugar per ha, but no clone showed a significant change in sugar per tonne of cane. The percentage losses of tonnes of sugar per ha were 4.58, 4.47, 4.41, and 3.96 for CP 70-1133, CP 72-1210, CP 74-2005, and CP 65-357 respectively.

A PROGRESS REPORT ON HARVESTING SYSTEM PERFORMANCE AT TWO FLORIDA SUGAR MILLS

B. R. Eiland U.S. Department of Agriculture Agricultural Research Service

Sugar Cane Production Systems Research Belle Glade, Florida

Field performance data were collected on hand and mechanical harvesting systems currently used by two Florida sugar mills. The study involved 46 fields of cultivar CP 72-1210, half of which were cut by hand and half by mechanical harvesters. Collected data included field conditions, harvesting losses, trash contents, and juice quality measurements for each field. The purpose of the study was to compare the field performance of the two harvesting systems and identify areas for improvement in each system. Recoverable sugar was lower with mechanical harvesting when compared to hand harvesting. Over half of the total sugar loss can be attributed to the difference in net cane between the harvesting systems. The remaining loss can be attributed to lower juice quality and undetermined or invisible losses.

104

PERFORMANCE OF THE NEW SUGAR CANE VARIETY CP 79-318 IN INFIELD VARIETY TESTS

Hugh P. Fanguy Sugar Cane Research Unit

Agricultural Research Service U.S. Department of Agriculture

Houma, Louisiana

The experimental yield data from infield variety tests of the new sugar cane variety CP 79-318, released for commercial production in 1987, are compared to the three major commercial varieties grown in Louisiana, CP 65-357,CP 70-321, and CP 74-383. This is the first opportunity to compare the yield performance of new varieties with the commercial standards when all plots are machine harvested. The preliminary results indicated that CP 79-318 compared favorably with commercial varieties in yield of sugar per hectare, tons of cane per hectare, sugar per ton, average weight per stalk and stalks per hectare. As a result of these tests, CP-318, as well as twelve varieties from this series, was introduced to the outfield tests and primary increase stations. However, only CP 79-318 was eventually released for commercial planting from this series.

THE ROLE OF STALK DENSITY, PITH AND TUBE IN SUGAR CANE SELECTION

K. A. Gravois and F. A. Martin Department of Agronomy

Louisiana Agricultural Experiment Station Louisiana State University Agricultural Center

Baton Rouge, Louisiana

A study determined the role of stalk density (g/cm3), pith and tube in a first line trial testing state. Eighty randomly selected sugar cane (Sacchamm supp.) clones were replicated three times in a randomized complete-block design. Data were collected in plant cane and first ratoon crops. Variances and covariances for the traits were estimated, and broad-sense heritabilities and genetic correlations were calculated. The genetic correlations were subjected to path-coefficient analysis to determine the relative effect of stalk density and tube on stalk weight (kg), and of pith on sucrose concentration (g sucrose/kg cane).

Broad-sense heritability is the extent to which an individual's phenotype is determined by its genotype. High broad-sense heritability indicates a potential for effective selection. The broad-sense heritabilities, on a singled plot basis, for stalk density, pith and tube were 0.028 ±0.005,0.612 ±0.096, and 0.423 ±0.070, respectively. Therefore, selection in a first line trial testing stage would be most effective for pith, tube, and stalk density, in that order.

Path-coefficient analysis was implemented to further define the genotypic correlations between trait relationships in a first ratoon crop. Stalk density and the tube were minor components of stalk weight, having direct effects of 0.170 and -0.005, respectively. The direct effect of pith on sucrose concentration was -0.005. The indirect effect of pith on sucrose concentration through its association with brix, a component of sucrose concentration, was -0.266 indicating that a negative relationship exists between pith and brix. Other results suggest that selection for clones with high brix, low pith and not tube indirectly improve stalk density.

105 105

SOME STUDIES ON DAMAGE TO SUGARCANE BY THE SPIDER MITE OUGONYCHUS STICKNEYI (McGREGOR)

David G. Hall Research Department, United States Sugar Corporation

Clewiston, Florida

Research was conducted during 1987 and 1988 on damage by the spider mite Oligonychus stickneyi (McGregor) to sugarcane in Florida. Estimates of population levels of mites on a per-square-centimeter basis were made on leaves with visible dewlaps using a hand magnifier fitted with a counting grid; multiple samples on leaves were taken to obtain an average level of mites/sq cm. In commercial fields, up to 50 nymphs and/or adults/sq cm and up to 273 eggs/sq cm were observed in individual samples, but overall levels usually averaged less than three mites/sq cm. Many mites were sometimes present on lower leaves while few were on upper leaves. Observations indicated that the russetting associated with mite infestations can occur where mites feed even when only several mites are present/sq cm. Based on tests with young sugarcane plants growing in containers, russetting was more pronounced in CL 61-620 and 'CL 59-1052 when ambient temperatures were cool (e.g., average daily temperature 68°F) than when they were warmer (e.g., average daily temperature 76°F) even though mite levels were similar; almost no russetting developed under greenhouse conditions when temperatures were high (e.g., average daily temperature 95°F). In an eight-week test that began soon after shoot emergence, shoot weights indicated growth of young CL 61-620 was reduced by 5 to 9% when mite levels averaged 1.7 nymphs and adults/sq cm and a moderate amount of russetting developed. No growth reductions occurred in two other eight-week tests with young plants of this variety when mite levels averaged 0.6 to 0.7 nymphs and adults/sq cm and little or no russetting developed. In a five-week greenhouse test with CP 74-2005 beginning four weeks after emergence, mite levels averaged 2.7 nymphs and adults/sq cm but almost no russetting occurred (average daily temperature 81°F); mites reduced the growth of primary shoots in this test by 13% based on shoot weights. At average mite levels of less than 1.0/sq cm and almost no russetting, no growth reductions were detected in young CL 59-1052, CP 70-1133, or CP 72-1210 shoots in an eight-week test beginning soon after shoot emergence. No reductions in growth occurred in young CL 59-1052 shoots in two eight-week tests that began soon after shoot emergence when mite levels averaged 1.0/sq cm and light to moderate russetting occurred.

ROLE OF VARIETIES, WEATHER CONDITIONS AND MANAGEMENT DECISIONS IN RECORD

SUGAR YIELDS FOR LOUISIANA IN 1987

Bengamin L. Legendre Sugar Cane Research Unit

Agricultural Research Service U.S. Department of Agriculture

Houma, Louisiana

Rigid selection for high sucrose content in the breeding and selection programs provided the varieties, and a combination of ideal weather conditions and management decisions provided the scenario for record sugar yields in 1987 of 110 kg/t (220 lb/t) of cane for the industry. During the period 1972 to 1987, the varieties CP 44-101, CP 48-103, CP 52-68, CP 61-37, and L 60-25 were replaced by the varieties CP 65-357, CP 70-321, CP 72-356, CP 72-370 and CP 74-384. Concurrently, sugar yields of maturity tests at the beginning of the harvest (October 1) increased from 59 kg/t (119 lb/t) of theoretical recoverable sugar per ton in 1972 to 105 kg/t (211 lb/t) in 1987. Further, end-of-harvest (December 1) sugar yields increased from 115 kg/t (230 lb/t) in 1972 to 142 kg/t (284 lb/t in 1987. Environmental or weather conditions that undoubtedly contributed to the record yields were rainfall distribution throughout the growing season and the intensity of sunlight prior to and during the harvest. Management decisions to use the chemical ripener glyphosate and new harvesting system, i.e., 2-row harvesters and chain pliers, also contributed to these record yields by enhancing maturity and reducing the amount of trash delivered to the mill. This paper will discuss the interaction of varieties, weather conditions, and management decisions, and attempt to explain how each of these factors contributed to the record sugar yields of 1987.

106

IMPROVING SUGAR CANE VARIETIES -WHAT ARE OUR OPTIONS?

F. A. Martin, S. B. Milligan, K. A. Gravois, and K. P. Bischoff Agronomy Department

Louisiana Agricultural Experiment Station Louisiana State University Agricultural Center

Baton Rouge, Louisiana

Improved sugar cane varieties are products of breeding program. The Louisiana Sugar Cane Variety Development Program uses a recurrent selection breeding strategy. This strategy returns the best offspring from one generation to be used as parents for the next generation. The genetic base for this program is kept broad through cooperation with the USDA interspecific breeding program at Houma, Louisiana and the USDA breeding program at Canal Point, Florida which incorporates worldwide commercial germplasm.

Means of improving genetic advance are explored. Genetic advance (GA) or the rate of genetic improvement per year is described by the formula: GA = ih2op/y where i is the selection intensity, h is the heritability, op is the phenotypic standard deviation and y is years per generation.

Earlier replicated testing across locations is the single best means to improve the rate of genetic advance. It simultaneously affects all elements contributing to genetic advance. Heritability is a function of genotypic, genotype by environment and error components earlier replicated testing across locations increases the accuracy of the genotype characterization. When the error and genotype by environment interaction components are decreased and the accuracy of genotypic characterization is improved, the heritability is increased. This facilitates earlier identification and return to the crossing program of superior genotypes. Earlier identification of elite parents translates into a shorter generation time and increased selection intensity.

INBREEDING IN THE LOUISIANA SUGAR CANE VARIETY DEVELOPMENT PROGRAM

AND THE UTILITY OF INBREEDING COEFFICIENTS AND PEDIGREES IN THE

VARIETY SELECTION PROCESS

Scott B. Milligan Agronomy Department

Louisiana Agricultural Experiment Station Louisiana State University Agricultural Center

Baton Rouge, Louisiana

Pedigrees were developed for sugar can genotypes in the Louisiana variety improvement program. A SAS computer program was written to provide a pedigree and inbreeding coefficient for any Louisiana variety and ancestors of interest through the 1986 assignments. The level of inbreeding for commercial Louisiana varieties ranged from a low of 3.6 per cent for CP 70-321 to 11.1 per cent for CP 65-357. Generally, the degree of inbreeding remained low but isolated cases of up to 30 per cent were identified. Although common ancestry of a few nobel canes in modern genotypes is widespread, pedigree examination made apparent the importance of new germplasm from the basic breeding program in preventing inbreeding.

Development of numerator relationship matrices using pedigrees and inbreeding coefficients is shown. Numerator relationship matrices incorporate genetic information of relatives in the selection process. The relationship matrices coupled with genetic variance-covariance matrices in mixed model analysis allows generation of best linear unbiased predictors (BLUP). These provide the most powerful value estimates of genotypes presently known.

107

INFLUENCE OF PROPICONAZOLE ON EMERGENCE OF CP 74-2005

Richard N. Raid Everglades Research and Education Center

Belle Glade, Florida

Propiconazole (Tilt) has been demonstrated to enhance emergence of sugar cane primary shoots by controlling pineapple disease (PD) and to increase germination even in the absence of the disease. Pineapple disease, caused by the fungus Ceratocystis paradoxa, is an important factor in establishment of new cane stands in many areas of the world. Although it has been reported in Florida, very little is known about its distribution or importance to the Florida sugar industry. Results of a greenhouse study examining varietal susceptibility to C. paradoxa indicate that at least two varieties, CP 72-2086 and CP 74-2005, appear to be very susceptible. These results may help to explain the poor germination often experienced with these particular varieties in commercial production fields.

A replicated field experiment was conducted to investigate the influence of propiconazole and a growth regulator (cytogen) on emergence of the PD susceptible variety CP 74-2005. Treatments consisted of an untreated check, propiconazole applied as a five minute cold water seedpiece dip, propiconazole applied as an in-furrow spray directed at the seedpiece prior to covering, and cytogen applied as an in-furrow directed spray prior to covering. The experiment was planted on January 8 in a poorly drained organic soil to favor disease development.

Examinations of excavated seedpieces indicated PD to be the primary cause of reduced stands. Propiconazole treatments resulted in stand counts which were significantly greater than those produced by the in-furrow spray application. Cytogen did not significantly increase emergence. These results suggest that propiconzole could be used to enhance emergence of PD susceptible varieties on Florida's organic soils.

SELECTION FOR SUGAR CANE BORER RESISTANCE IN SEEDLING PROGENIES

W. H. White, J. W. Dunckelman, and B. L. Legendre Sugar Cane Research Unit

Agricultural Research Service U.S. Department of Agriculture

Houma, Louisiana

Sugar cane clones identified as resistant to the sugar cane borer (SCB), Diatraea saccharalis (F.) were used to make bi-parental crosses during the 1986 crossing campaign at Canal Point, Florida. Among 11 bi-parental crosses made were intercrosses of SCB resistant 1983 CPs, SCB resistant 1983 CPs x commercial varieties, and commercial varieties x commercial varieties.

A total of 5,174 seedlings were transplanted to the field during the spring of 1987 in a 4 to 1 skip-row pattern. This pattern consisted of four rows of cane followed by one row of corn used as the inoculated host. The field and surrounding headland was then sprayed with chlorpyrifos to suppress fire ant populations. Just prior to tassling, the corn was artificially infested with approximately 10 (±2) laboratory-reared first-instar SCB by means of a hand-held inoculator. Sugar cane seedlings were selected September 25 in the plant cane crop, selecting first for SCB resistance and second for agronomic type; i.e., numbers of stalks, length and diameter of stalks, and general vigor. No attempt was made to select for brix. A total of 335 seedlings were advanced to clonal plots which will be treated similarly. Selection rate ranged from 0.9 per cent to 12.2 per cent. The cross CP 74-383 x CP 72-356 (susceptible x susceptible), included as a check, had a selection rate of 2.6 per cent.

108

ABSTRACTS - MANUFACTURING

TURBIDIMETRIC EVALUATION OF NOVEL CLARIFICATION SCHEMES AND EVAPORATION

G. A. Adongo Department of Food Science, Cornell University

Ithaca, New York

S. J. Clarke Audubon Sugar Institute

Louisiana Agricultural Experiment Station Louisiana State University Agricultural Center

Baton Rouge, Louisiana

Turbidimetric measurements can be very informative in studies of sugar factory performance, in particular for clarifier operation and raw sugar quality. Techniques for determination of turbidity have been cumbersome, but modern instruments can greatly reduce the time and effort involved. The refractive index (and therefore the brix) of the solution has a major effect on the measured value of turbidity.

The residual turbidity of clarified juice is a useful measure of the removal of suspended material. Several novel clarification schemes involving anionic and cationic polymers and surfactants have been studied. Cationic polymers can be very effective in reducing turbidity and comparative data will be given for several chemical treatments of juice. Evaporation of juice to syrup causes formation of suspended material, e.g., the calcium salts related to scaling. Changes in turbidity through the evaporator were measured on pilot scale equipment at ASI and at three mills. These and results for other factory streams are described.

PARAMETERS FOR VACUUM PAN AUTOMATION

G. L. Aleman, Retired Sugar Cane Growers Cooperative of Florida

Belle Glade, Florida

The success or failure in the automatic operation of any process equipment depends mostly in the proper selection of the parameters to be monitored and/or controlled.

The purpose of this paper is to try to highlight those parameters essential in the process of massecuite boiling in vacuum pans, and to develop the necessary communication between the experienced people in the art of sugar boiling and the personnel with knowledge of the instrument equipment and resources that modern technology has made available.

SHORT TERM/LONG TERM COMPUTER APPLICATION FOR THE SUGAR INDUSTRY

E. Alfonso and R. Valdes Okeelanta Corporation

South Bay, Florida

As in many industries, for the past 20 years the introduction of computers has been viewed with mixed approaches. No doubt, since the wheel; it is the most revolutionary change introduced, at the fastest speed, with the widest of applications.

The use of programmable controls has been quickly accepted throughout industry worldwide. The ability to change a control scheme quickly and without component replacement is a very desirable characteristic of these

109

systems. With the addition of "smart" components capable of interactive control through the use of communication paths, a wider field has opened up. These pathways allow for "soft" wiring between intelligent systems which can be quickly reconfigured to allow for any change in even the most complex scheme without the need to rewire a component.

Economically this is a welcome advantage, for a technician can rapidly modify a system with a configuration terminal without ever having to make a physical change, saving on down time and labor expense. In many industries computer systems have been installed as a matter of survival, in the fastest and most developing technological explosion time in history.

EVALUATIONS OF THE PERFORMANCE OF A FORCED FEED ROLLER ON THE SEVENTH MILL AT ATLANTIC SUGAR ASSOCIATION

J. F. Alvarez, H. Cardentey, and A. Pacheco Atlantic Sugar Association

Belle Glade, Florida

The performance of the Forced Feed Roller is measured against other crops in terms of imbibition and percentage of pol in bagasse. Theoretical performance is compared to actual performance. The benefits of the Forced Feed Roller are evaluated as well as the problems during the operation.

COMPUTER MODEL TO ASSESS THE ECONOMIC VALUE OF A SUGAR CANE VARIETY

S. J. Clarke and S. B. Milligan Audubon Sugar Institute and Agronomy Department

Louisiana Agricultural Experiment Station Louisiana State University Agricultural Center

Baton Rouge, Louisiana

A computer model was developed to estimate sugar and molasses output from a typical sugar mill for non-standard cane varieties. The program, written in the BASIC and SAS computer languages, assumes the operating conditions for a standard cane variety: a fixed mill fiber throughput, fixed evaporator load and boiling house efficiency, and values for syrup brix, molasses brix and fiber imbibition per cent. The standard cane composition may be from commercial operations or from the breeding program. The model enables comparison of varieties with different fiber, juice brix and juice purity levels in terms of sugar and molasses production per ton of cane per day. The program used in conjunction with relative harvesting, transportation and factory costs allows computation and comparison of the economic value of different varieties.

CROWN WHEEL REMOVAL FROM BAGASSE ROLL

G. Delaune, and J. Theriot Breaux Bridge Sugar Cooperative

Breaux Bridge, Louisiana

Information is reviewed and compiled on the effects of milling without crown wheels on the bagasse rolls. Discussed are details on performance before and after removal of the crown wheels at Breaux Bridge Sugar Cooperative.

110

MICROPROCESSOR CONTROL STRUCTURES FOR RAW SUGAR FACTORIES

W. Keenliside Audubon Sugar Institute

Louisiana Agricultural Experiment Station Louisiana State University Agricultural Center

Baton Rouge, Louisiana

The implementation of digital control in sugar factories requires an understanding of the structures and architecture available for control systems. This paper describes the available methods for control systems and shows the different areas within a factory which can be linked together to provide both process control and operational decision making. Methods of constructing a hierarchy for the control equipment is developed and different options are presented.

MONOCAST NYLON MILL BEARING LINERS

K. McGrew Audubon Sugar Institute

Louisiana Agricultural Experiment Station Louisiana State University Agricultural Center

Baton Rouge, Louisiana

Results and analysis of plastic inserts installed on quarterbox bearings of the bagasse roll at Breaux Bridge Sugar Cooperative. Industrial plastics are becoming more popular in areas of intensive wear, heat, or corrosion. Evaluations and assessments are based on one crop year. Further results will be accumulated during the 1988 harvest.

IMPROVING PERFORMANCE OF LOW-GRADE CRYSTALLIZERS

Y. Oubrahim, and M. Saska Audubon Sugar Institute

Louisiana Agricultural Experiment Station Louisiana State University Agricultural Center

Baton Rouge, Louisiana

M. Garcia St. James Sugar Cooperative

St. James, Louisiana

Various policies of operating the low-grade crystallizers and their effects on molasses exhaustion are discussed. Comparison is made between continuous versus batch operation and three ways to reduce the massecuite viscosity, i.e. dilution with water, dilution with molasses and temperature control with no dilution, are evaluated in terms of their effects on sucrose loss. The factors limiting the flow of heavy massecuites, such as the elevation between crystallizers and the cross-sectional area of the connectors are briefly discussed.

CANE KNIVES CHOKE PROTECTION

A. L. Perera Okeelanta Corporation

South Bay, Florida

A description is given of a new approach for the protection of the cane knives and cane carriers against chokes due to cane overfeeding, rocks, or any large piece of metal, chain, etc. carried along with the cane. The system consists of a set of electronic speed switches that could be set at a desired speed according to the

111

particular conditions or preferences of each installation. A magnetic pick-up senses the signal from a small spur gear (previously attached to the free end of the cane knives shaft), and sends it to the speed switch, through a digital tachometer for a continuous readout of knife speed and an easy set up.

PREPARING CANE WITH AN ELECTRONIC GOVERNOR

L. R. Zarraluqui Sugar Cane Growers Cooperative of Florida

Belle Glade, Florida

A second-hand two-stage, condensing steam turbine without governor, built in 1950, was installed in the Okeelanta Sugar Factory, replacing a wrecked turbine that used to drive the first set of cane cutting knives of a 12,000 TCD tandem. Due to major differences between design and local conditions, the reapplication needed rerating the turbine through an engineering study. Also, among several options, a solid-state electronic governor was selected as a retrofit for the rerate turbine.

The governor proved to be extremely precise, and to be capable of providing unsurpassed stability to the prime move. Moreover, it happens to be remarkable reliable, too. Installation took place in the summer of 1984, during the repair season. Today, four harvests and some six million tons of cane later, there has been no downtime due to governor failure, notwithstanding the fact that the governor requires little, if any, maintenance. This paper describes the rerate, the governor and the governor characteristics.

112

AMERICAN SOCIETY OF SUGAR CANE TECHNOLOGISTS EDITORIAL POLICY

Nature of papers to be published:

Papers submitted must represent a significant technological or scientific contribution. Papers will be limited to the production and processing of sugarcane, or to subjects logically related. Authors may submit papers that represent a review, a new approach to field or factory problems, or new knowledge gained through experimentation. Papers promoting machinery or commercial products will not be acceptable.

The Journal will appear at least once a year. At the direction of the Joint Executive Committee, the Journal may appear more frequently. Contributed papers not presented at a meeting may be reviewed, edited, and published if the editorial criteria are met.

Editorial Committee:

The Editorial Committee shall be composed of the managing editor, technical editor for the Agricultural Section and technical editor for the Processing Section.

The Editorial Committee shall regulate the Journal content and assure its quality. They are charged with the authority necessary to achieve these goals. The Editorial Committee shall determine broad policy. Each editor will serve for three years; he may at the Joint Executive Committee's discretion, serve beyond the expiration of his term.

Handling of manuscripts:

Four copies of each manuscript are submitted to the managing editor. Manuscripts received by the managing editor will be assigned a registration number determined serially by the date of receipt. The managing editor writes to the one who submitted the paper to inform the author of the receipt of the paper, the registration number which must be used in all correspondence regarding it, and the page cost of publishing.

The technical editor receives from the managing editor all papers whose subject matter falls in his "area." He obtains at least two reviews for each paper from qualified persons. The identities of reviewers must not be revealed to each other nor to the author during the review process. Instructions sent with the papers emphasize the necessity for promptness as well as thoroughness in making the review. Page charges will be assessed for the entire manuscript for non-members. Members will be assessed for those pages in excess of ten (10) double spaced pica typed pages of 8 1/2" x 11" dimension with one (1) inch margins.

When a paper is returned by a reviewer, the technical editor evaluates the paper and the recommendations of the reviewers. If the paper as received is recommended by two reviewers for publication in the Journal, it is sent to the managing editor.

If major revisions are recommended, the technical editor sends the paper to the author for this purpose, along with anonymous copies of reviewers' recommendations. When the paper is returned to the technical editor, he will judge the adequacy of the revision and should send the paper back to any reviewer who requested major changes, for his further review. When the paper has been revised satisfactorily, it is sent to the managing editor for publishing. A paper sent to its author for revision and held more than 6 months will be given a new date of receipt when returned. This date will determine the priority of publication of the paper.

A paper rejected by one reviewer may be sent to additional reviewers until two reviewers either accept or reject the paper.

If a paper is judged by two or more reviewers as not acceptable for the Journal, the technical editor returns it to the author along with a summary of the reasons given by the reviewers for the rejection. The registration form for the paper is filled out and returned to the managing editor along with copies of the

113

reviewers' statements and a copy of the technical editor's transmittal letter to the author. The reviewers' statements should not be forwarded to the author in this instance.

The names of all reviewers must be shown on the registration form.

After the review process is completed, each accepted paper is read by the technical editor to correct typographical, grammatical, and style errors and to improve the writing where this seems possible and appropriate, with special care not to change the meaning. Instructions for the printer are inserted as needed. The papers are sent by the technical editor to the managing editor who notifies the authors of this fact and of the probable dates of publication.

Preparation of capers for publication:

Papers sent by the technical editor to the managing editor are prepared for printing according to their dates of original submittal and final approval and according to the space available in the next issue of the Journal.

Tables are retyped in the proper form for reproduction, and proofs are sent to the authors along with the galley proofs. When the proofs are returned, all necessary corrections are made prior to reproduction.

The drawings and photographs for the figures in the paper are "scaled" according to their dimensions, the size of lettering, and other factors. They are then sent to the printer for camera work. Proofs of the illustrations are sent to the authors. Any changes requested at this stage would be expensive and authors will be expected to pay the cost of such changes.

The author will be notified at the appropriate time that he may order reprints at cost.

Reprinting in trade journals has the approval of the Editorial Committee provided: a) no article is reprinted before being accepted by the Journal; b) credit is given the author, his institution and the ASSCT; and c) permission of the author has been obtained. Summaries, condensations, or portions may be printed in advance of Journal publication provided the approval of the Editorial Committee has been obtained.

114

RULES FOR PREPARING PAPERS TO BE PRINTED IN THE JOURNAL OF THE AMERICAN SOCIETY OF SUGAR CANE TECHNOLOGISTS

Format

Unless the nature of the manuscript prevents, it should include the following sections in the order listed: ABSTRACT, INTRODUCTION, MATERIALS and METHODS, RESULTS, DISCUSSION, CONCLUSIONS, ACKNOWLEDGMENTS, and REFERENCES. Not all the sections listed above will be included in each paper, but each section should have an appropriate heading that is centered on the page with all letters capitalized.

Authorship

Name of the author(s), institution or organization with which he is associated, and the location should follow the title of the paper.

Abstract

The abstract should be placed at the beginning of the manuscript, immediately following the author's name, organization and location.

Tables

Number the tables consecutively and refer to them in the text as Table 1, Table 2, etc. Each table must have a heading or caption. Capitalize only the initial word and proper names in table headings. Headings and text of tables should be single spaced. Each table should be on a separate sheet.

Drawings & Photographs

Drawings and photographs must be provided separately from the text of the manuscript. Type figure numbers and legends on separate pieces of paper with proper identification. Drawings and photographs should be of sufficient quality that they will reproduce legibly.

Reference Citations

The heading for the literature cited should be REFERENCES. References should be arranged such that the literature cited will be numbered consecutively and placed in alphabetical order according to the surname of the senior author. In the text, references to literature cited can be made by number or name of author and number from list of references. (See example.) Do not use capital letters in the titles of such articles except in initial words and proper names, but capitalize words in the titles of the periodicals or books.

Suggested Format (Examples below)

EVALUATION OF SUGARCANE CHARACTERISTICS FOR MECHANICAL HARVESTING IN FLORIDA

J. E. Clayton and B. R. Eiland Agricultural Engineers, SEA, USDA, Belle Glade, Florida

J. D. Miller and P. Tai Research Geneticists, SEA, USDA, and Canal Point, Florida

ABSTRACT

INTRODUCTION

MATERIALS AND METHODS

115

RESULTS

Table 1. Varietal characteristics of nine varieties of sugarcane over three-year period at Belle Glade, Florida.

Figure 1. Relative size of membrance pores.

DISCUSSION

CONCLUSIONS

ACKNOWLEDGMENTS

REFERENCES

1. Arceneaux, G. 1935. A simplified method of making theoretical sugar yield calculations in accordance with Winter-Carp-Geerligs formula. Intnl. Sugar Jour. 37:264-265.

2. Florida Sugar Cane League, Inc. 1978. Florida's Sugar Industry Brochure distributed by the Florida Sugar Cane League, Inc., Clewiston, Florida.

3. Gascho, G. J., J. E. Clayton, and J. P. Gentry. 1973. Sugarcane deterioration during storage as affected by chopping, delay in milling, and burning. Proc. ASSCT 2(NS): 168-172.

4. Steel, R. G. D. and J. H. Torrie. 1960. Principles and Procedures of Statistics. McGraw-Hill Book Co., Inc., N. Y.

116

AUTHOR INDEX

Adongo.G. A 109 Harrison, N. A. 104 Alfonso, E. 109 Hoy, J. W. 6 Aleman, G. L. 109 Keenliside W. I l l Alvarez, J. F. 110 Legendre, Benjamin L 106, 108 Anderson, David L. 44 Martin, F. A. 6, 17, 105, 107 Arceneaux, Allen 103 McGrew, K. I l l Bishoff, Keith P. 17, 107 Miller, J. D 30, 62 Bourg, Druis 97 Millhollon, R. W 91 Breithaupt, J. A. 103 Milligan, S. B 107, 110 Cardentey, H. 110 Oubrahim, Y. I l l Chao, C. P. 6 Pacheco,A 110 Chapman, Brian A 22 Parrado, R 71 Cherry, Ron H. 52,103 Perera, A. L. I l l Chew, Victor 30 Quebedeaux, Joey P 17 Christy, Ralph D. 22 Raid, Richard N 108 Clarke, S. J. 109, 110 Ricaud, Ray 103 Coale, F. J. 52 Richard, Edward P., Jr 38 Davis, M. J. 104 Rozeff, N 81 Dean, J. L. 104 Saska, M I l l Delaune, G. 110 Shine, James M., Jr 30 Dunckelman, J. W. 56,108 Tai, P. Y. P 62 Eiland, B. R. 104 Theriot, J. 110 Fanguy, Hugh P. 91, 105 Thomas, J. R 81 Garcia, M. I l l Ulloa, M. F 44,71 Glaz, Barry 71 Viator, Howard P 38 Gravois, K. A. 105, 107 Valdes, R 109 Grisham, Michael P. 97 White, W. H. 56, 108 Hall, David G 106 Zarraluqui, L. R 112

117